CA2164489A1 - Traffic management and congestion control for packet-based networks - Google Patents
Traffic management and congestion control for packet-based networksInfo
- Publication number
- CA2164489A1 CA2164489A1 CA002164489A CA2164489A CA2164489A1 CA 2164489 A1 CA2164489 A1 CA 2164489A1 CA 002164489 A CA002164489 A CA 002164489A CA 2164489 A CA2164489 A CA 2164489A CA 2164489 A1 CA2164489 A1 CA 2164489A1
- Authority
- CA
- Canada
- Prior art keywords
- packets
- virtual connection
- information rate
- committed
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/82—Miscellaneous aspects
- H04L47/822—Collecting or measuring resource availability data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/0289—Congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L12/5602—Bandwidth control in ATM Networks, e.g. leaky bucket
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/11—Identifying congestion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/26—Flow control; Congestion control using explicit feedback to the source, e.g. choke packets
- H04L47/263—Rate modification at the source after receiving feedback
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/28—Flow control; Congestion control in relation to timing considerations
- H04L47/283—Flow control; Congestion control in relation to timing considerations in response to processing delays, e.g. caused by jitter or round trip time [RTT]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/35—Flow control; Congestion control by embedding flow control information in regular packets, e.g. piggybacking
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/76—Admission control; Resource allocation using dynamic resource allocation, e.g. in-call renegotiation requested by the user or requested by the network in response to changing network conditions
- H04L47/765—Admission control; Resource allocation using dynamic resource allocation, e.g. in-call renegotiation requested by the user or requested by the network in response to changing network conditions triggered by the end-points
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/70—Admission control; Resource allocation
- H04L47/82—Miscellaneous aspects
- H04L47/826—Involving periods of time
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/0284—Traffic management, e.g. flow control or congestion control detecting congestion or overload during communication
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/10—Flow control between communication endpoints
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5614—User Network Interface
- H04L2012/5615—Network termination, e.g. NT1, NT2, PBX
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5619—Network Node Interface, e.g. tandem connections, transit switching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5629—Admission control
- H04L2012/5631—Resource management and allocation
- H04L2012/5632—Bandwidth allocation
- H04L2012/5635—Backpressure, e.g. for ABR
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5629—Admission control
- H04L2012/5631—Resource management and allocation
- H04L2012/5636—Monitoring or policing, e.g. compliance with allocated rate, corrective actions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/54—Store-and-forward switching systems
- H04L12/56—Packet switching systems
- H04L12/5601—Transfer mode dependent, e.g. ATM
- H04L2012/5638—Services, e.g. multimedia, GOS, QOS
- H04L2012/5646—Cell characteristics, e.g. loss, delay, jitter, sequence integrity
- H04L2012/5652—Cell construction, e.g. including header, packetisation, depacketisation, assembly, reassembly
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/50—Reducing energy consumption in communication networks in wire-line communication networks, e.g. low power modes or reduced link rate
Abstract
In a packet-based communications network (100), a virtual connection is established between a source node (102a) and a destination node (102b), which may traverse one or more intermediate nodes (105). During the flow of packets along the virtual connection in a forward direction, from the source node (102a) to the destination node (102b), each node measures the utilization of critical resources (CUF), and this utilization information is inserted into return packets flowing in the backward (return) direction, from the destination node (102b) to the source node (102b). CUF is indicative of the maximum utilization of any resource in the virtual connection forward path. If the network utilization information indicates that the resources of a virtual connection are under-utilized, the submission rate (SIR) of packets onto the virtual connection is increased.
However, if the network utilization information indicates that the resources of the virtual connection are over-utilized, the submission rate of packets onto the network at the source node is reduced.
However, if the network utilization information indicates that the resources of the virtual connection are over-utilized, the submission rate of packets onto the network at the source node is reduced.
Description
- ~'~64~89 Traffic Management and Congestion Control for F~acket-Based Networks Technical Field:
The presenit invention relates to packet-based networks, and more particularly, to traffic management and congestion control for packet-based networks.
l0 Backaround of tile Invention:
A packet-b<~sed communications network provides for the transmission of packets between various nodes in the communications network. A "packet" comprises any fixed or variable sized grouping of bits. Examples of packet-based networks _Lnclud~e: frame relay networks wherein a packet corresponds to a frame relay frame; a cell-switching network, e.g., an asynchronous transfer mode 20 (ATM) network, wherein a packet corresponds to an ATM
cell; etc. A node ma,y be defined as a termination point for two or more channels or lines, and it typically includes contro.'L equipment.
Within in ~~ packet-based communications network, a virtual connectpon (VC) is established between a source node and a destp_nation node, and usually traverses one or more intermediate nodes. A virtual connection is a logical connection (channel) between the source node and 30 the destination node :for the transmission of packets therebetween. Each node in the virtual connection acts as a packet exchanger for receiving and forwarding a packet to its destination node. Each node within the communications network may be part of a variety of virtual connections. Terminal nodes, such as the source 21fi4~89 node and the destination node, have the necessary interface cards for receiving preformatted packets, and/or properly formatting data contained in a standard user data stream into packets, and/or for reassembling packets into the standard user data streams.
Packet-bas~sd communication networks are able to integrate a variety of data such as local area network data, voice data, and image/video data into packets for transmission over high-speed digital trunks. Each source node and destin;~tion node is interconnected to various customer premis~ss equipment (CPE) such as a local area network, fax/modem equipment, voice equipment, and image/video equipment and other data and/or packet equipment for transmission of corresponding packets via the communications network to a destination node.
In such packet-based networks, it is important to provide appropriate protocols for traffic management and congestion control of packets traveling via a virtual connection. Various 'methods have been developed to control the rates of packet transmission onto a virtual connection and also control the selective discarding of packets for pac)cet management and congestion control.
One approach to :packet management and congestion control is a crEadit based approach. The credit based approach is a protocol which acts between adjacent nodes, identified as a transmitter and a receiver, wherein a packet is not transmitted from a transmitter to a receiver unless the sender knows that the receiver can accept the information without loss. The receiver sends control informai:ion to the transmitter indicating that it ~~~44~9 is able to accept new packets. The credit based approach does not rely on end-to-end signals to respond to congestion. In such a credit based approach, closed loop control of the 'transmitter is utilized so that data is sent as fast as possible, with no loss of data, from the transmitter to the receiver.
Another congestion management approach is a feedback approach wherein the congestion state of the virtual connection is measured as packets flow in a forward direction from i;.he source node to the destination node, and the measurement of the congestion state is sent back from the destin<~tion node to the source node for adjusting the submission of packets onto the network.
Summary of the =Cnvention:
Objects of the invention include improved traffic management and congestion control of a packet-based communications network.
Another ob=ject of the present invention is a packet-based communicai:ions ;network wherein the data rate of a virtual connect:lon is controlled on an end-to-end basis.
A further abject of the present invention is traffic management of a packet-based communications network wherein utilization o:f network resources is maximized while over utilization of network resources i.s limited.
A still further object of the present invention is to provide a packet-based communications network having an initial packE~t submission rate which is selected to 2184489 .-limit delays incurred by bursts in traffic while also minimizing congestion at convergence points of the network.
Another object of the present invention is to provide a packet-based communication network having a quick and fair allocation of resources among active virtual connections.
A still further object of the present invention is the timely provision of network utilization information to a source for adjustment of the rate of packet submission by the source, with the most timely utilization information being related to a critical or access node of a virtual connection.
According to t:he present invention, there is provided a system for traffic management and congestion control on a virtual connection of a packet-based communications network, comprising:
customer premisea equipment for providing packets for transmission on t:he virtual connection, and for receiving packEas transmitted on the virtual connection;
a source node for accepting packets from corresponding :source node customer premises equipment, and for transm:Ltting the packets over the virtual connection through the packet-based communications network at a variable submission information rate (SIR);
a destination node for accepting the packets transmitted in a forward direction over the virtual connection from said source node to said destination node, for providing the packets transmitted in the forward direction to corresponding destination node customer premises equipment connected thereto, and for transmitting return packets in a return direction over the virtual connection from said destination node to said source node; a.nd r z 21 64~ 489 means for measuring a channel utilization factor (CUF), which is directly related to the degree of utilization of network resources, during transmission of packets in said forward direction, and for inserting said channel utilization factor in return packets travelling in said return direction;
wherein said source node comprises means for controlling said submission information rate in response to said channel utilization factor contained in said return packets.
According to the present invention there is also provided a method for traffic management and congestion control of packets traveling in a forward direction from a source node to a destination node on a virtual connection of a packet-based communications network, the method comprising the steps of:
transmitting the packets from the source node to the destination node on the virtual connection at a variable submission information rate (SIR);
measuring a channel utilization factor (CUF), which is directly re:Lated i~o the degree of utilization of network resources, during transmission of packets in said forward direction;
inserting said channel utilization factor in return packets travelling in a return direction from the destination node to 'the source node on the virtual connection; and adjusting said submission information rate in response to said channel utilization factor contained in said return packets.
According to the present invention, in a packet-based communications network, a virtual connection is established between a source node and a destination node, which may traverse one or more intermediate nodes; during 4a sz...
the flow of pac;kets a long the virtual connection in a forward direction, from the source node to the destination node, each node measures the utilization of critical resources, and this utilization information is piggy-backed onto pac:kets flowing in the backward (return) direction, i:rom the destination node to the source node.
In further accord with the present invention, the utilization information provided to the source node is indicative of t:he ma~cimum utilization of any resource in the virtual connection forward path.
4b ~1s4489 In still f~srther accord with the present invention, in the absence of packets flowing in the backward direction, the destination node generates special empty packets for transmission to the source node upon which the utilization information may be piggy-backed by the intermediate nodes.
According :Further to the present invention, if the network utilization information indicates that the resources of a ~~irtual connection are under-utilized, the submission rate of packets onto the virtual connection is increased. How~wer, if the network utilization information indicates that the resources of the virtual connection are over-utilized, the submission rate of packets onto thEa network at the source node is reduced.
According :still further to the present invention, the period of tame between two consecutive rate adjustments of i~he packet submission rate of packets onto a virtual connecaion is selected to allow the first adjustment to take effect throughout the network before the occurrence of the second adjustment.
According still further to the present invention, when a virtual connection becomes active, e.g., a source node begins transmission of packets onto the virtual connection, othsar active virtual connections reduce their use of excess nEatwork resources in favor of the newly active virtual connection, thereby allowing a quick and fair allocation of resources among active virtual connections. Additionally, when network utilization information ind=Lcates that a reduction in the rate of packets submittE~d onto the network is necessary, virtual ~1.644~9 connections with higher use of excess network resources incur higher packet submission rate reductions compared with virtual connections with lesser use of excess network resources.
In further accord with the present invention, bandwidth admis:ion for a virtual connection is controlled in accordance with assigned service categories for the determination of whether or not packets are admitted to the network, and for the identification of packets that arEa admitted to the network with an assigned priority for dei~ermin,ation of which packets will be discarded under certain congestion conditions.
In still further accord with the present invention, credits are associated with each service category such that various crE~dits ~3re assigned to a virtual connection per measurement period, T, for admission of packets to the network durp_ng the measurement period. Each packet is assigned a credit, and during extreme congestion conditions on the network, packets are discarded based on the priority of the corresponding assigned credit.
In further accord with the present invention, multiple classe~~ of scarvice may be designated which define the bandwidth admission, trunk queue and service discipline, and CUF measurement and SIR control for a given virtual connection.
The foregoing and other objects, features and advantages of tl~~e pre~~ent invention will become more apparent in light of i;he following detailed description X1.64489 of exemplary embodiments thereof as illustrated in the accompanying drawings.
Brief Descripti~~n of the Drawing's:
Fig. 1 is ;a schematic block diagram of a packet-based communications network;
Fig. 2 is ~~ schematic block diagram of an access module containers within a source node or a destination node of the communications network of Fig. 1;
Fig. 3 is ~~ schematic block diagram of a transit module contained within a node of the communications network of Fig. 1;
Fig. 4 is <3 diagram showing a packet having a header portion and a d<~ta portion;
Fig. 5 is ~~ grap:h depicting the measurement of one network round-trip delay (RTD) of a packet in the communications network; and Fig. 6 is ~~ graplh illustrating the relationship between rate cot~tro~.,and channel utilization factor (CUF) .
Best Mode for Carrying Out the Invention:
For purposE~s of simplifying the description of the present invention, number abbreviations are used in the following descrp_ption. The following table of abbreviations i~~ prov:ided as a reference for the reader:
The presenit invention relates to packet-based networks, and more particularly, to traffic management and congestion control for packet-based networks.
l0 Backaround of tile Invention:
A packet-b<~sed communications network provides for the transmission of packets between various nodes in the communications network. A "packet" comprises any fixed or variable sized grouping of bits. Examples of packet-based networks _Lnclud~e: frame relay networks wherein a packet corresponds to a frame relay frame; a cell-switching network, e.g., an asynchronous transfer mode 20 (ATM) network, wherein a packet corresponds to an ATM
cell; etc. A node ma,y be defined as a termination point for two or more channels or lines, and it typically includes contro.'L equipment.
Within in ~~ packet-based communications network, a virtual connectpon (VC) is established between a source node and a destp_nation node, and usually traverses one or more intermediate nodes. A virtual connection is a logical connection (channel) between the source node and 30 the destination node :for the transmission of packets therebetween. Each node in the virtual connection acts as a packet exchanger for receiving and forwarding a packet to its destination node. Each node within the communications network may be part of a variety of virtual connections. Terminal nodes, such as the source 21fi4~89 node and the destination node, have the necessary interface cards for receiving preformatted packets, and/or properly formatting data contained in a standard user data stream into packets, and/or for reassembling packets into the standard user data streams.
Packet-bas~sd communication networks are able to integrate a variety of data such as local area network data, voice data, and image/video data into packets for transmission over high-speed digital trunks. Each source node and destin;~tion node is interconnected to various customer premis~ss equipment (CPE) such as a local area network, fax/modem equipment, voice equipment, and image/video equipment and other data and/or packet equipment for transmission of corresponding packets via the communications network to a destination node.
In such packet-based networks, it is important to provide appropriate protocols for traffic management and congestion control of packets traveling via a virtual connection. Various 'methods have been developed to control the rates of packet transmission onto a virtual connection and also control the selective discarding of packets for pac)cet management and congestion control.
One approach to :packet management and congestion control is a crEadit based approach. The credit based approach is a protocol which acts between adjacent nodes, identified as a transmitter and a receiver, wherein a packet is not transmitted from a transmitter to a receiver unless the sender knows that the receiver can accept the information without loss. The receiver sends control informai:ion to the transmitter indicating that it ~~~44~9 is able to accept new packets. The credit based approach does not rely on end-to-end signals to respond to congestion. In such a credit based approach, closed loop control of the 'transmitter is utilized so that data is sent as fast as possible, with no loss of data, from the transmitter to the receiver.
Another congestion management approach is a feedback approach wherein the congestion state of the virtual connection is measured as packets flow in a forward direction from i;.he source node to the destination node, and the measurement of the congestion state is sent back from the destin<~tion node to the source node for adjusting the submission of packets onto the network.
Summary of the =Cnvention:
Objects of the invention include improved traffic management and congestion control of a packet-based communications network.
Another ob=ject of the present invention is a packet-based communicai:ions ;network wherein the data rate of a virtual connect:lon is controlled on an end-to-end basis.
A further abject of the present invention is traffic management of a packet-based communications network wherein utilization o:f network resources is maximized while over utilization of network resources i.s limited.
A still further object of the present invention is to provide a packet-based communications network having an initial packE~t submission rate which is selected to 2184489 .-limit delays incurred by bursts in traffic while also minimizing congestion at convergence points of the network.
Another object of the present invention is to provide a packet-based communication network having a quick and fair allocation of resources among active virtual connections.
A still further object of the present invention is the timely provision of network utilization information to a source for adjustment of the rate of packet submission by the source, with the most timely utilization information being related to a critical or access node of a virtual connection.
According to t:he present invention, there is provided a system for traffic management and congestion control on a virtual connection of a packet-based communications network, comprising:
customer premisea equipment for providing packets for transmission on t:he virtual connection, and for receiving packEas transmitted on the virtual connection;
a source node for accepting packets from corresponding :source node customer premises equipment, and for transm:Ltting the packets over the virtual connection through the packet-based communications network at a variable submission information rate (SIR);
a destination node for accepting the packets transmitted in a forward direction over the virtual connection from said source node to said destination node, for providing the packets transmitted in the forward direction to corresponding destination node customer premises equipment connected thereto, and for transmitting return packets in a return direction over the virtual connection from said destination node to said source node; a.nd r z 21 64~ 489 means for measuring a channel utilization factor (CUF), which is directly related to the degree of utilization of network resources, during transmission of packets in said forward direction, and for inserting said channel utilization factor in return packets travelling in said return direction;
wherein said source node comprises means for controlling said submission information rate in response to said channel utilization factor contained in said return packets.
According to the present invention there is also provided a method for traffic management and congestion control of packets traveling in a forward direction from a source node to a destination node on a virtual connection of a packet-based communications network, the method comprising the steps of:
transmitting the packets from the source node to the destination node on the virtual connection at a variable submission information rate (SIR);
measuring a channel utilization factor (CUF), which is directly re:Lated i~o the degree of utilization of network resources, during transmission of packets in said forward direction;
inserting said channel utilization factor in return packets travelling in a return direction from the destination node to 'the source node on the virtual connection; and adjusting said submission information rate in response to said channel utilization factor contained in said return packets.
According to the present invention, in a packet-based communications network, a virtual connection is established between a source node and a destination node, which may traverse one or more intermediate nodes; during 4a sz...
the flow of pac;kets a long the virtual connection in a forward direction, from the source node to the destination node, each node measures the utilization of critical resources, and this utilization information is piggy-backed onto pac:kets flowing in the backward (return) direction, i:rom the destination node to the source node.
In further accord with the present invention, the utilization information provided to the source node is indicative of t:he ma~cimum utilization of any resource in the virtual connection forward path.
4b ~1s4489 In still f~srther accord with the present invention, in the absence of packets flowing in the backward direction, the destination node generates special empty packets for transmission to the source node upon which the utilization information may be piggy-backed by the intermediate nodes.
According :Further to the present invention, if the network utilization information indicates that the resources of a ~~irtual connection are under-utilized, the submission rate of packets onto the virtual connection is increased. How~wer, if the network utilization information indicates that the resources of the virtual connection are over-utilized, the submission rate of packets onto thEa network at the source node is reduced.
According :still further to the present invention, the period of tame between two consecutive rate adjustments of i~he packet submission rate of packets onto a virtual connecaion is selected to allow the first adjustment to take effect throughout the network before the occurrence of the second adjustment.
According still further to the present invention, when a virtual connection becomes active, e.g., a source node begins transmission of packets onto the virtual connection, othsar active virtual connections reduce their use of excess nEatwork resources in favor of the newly active virtual connection, thereby allowing a quick and fair allocation of resources among active virtual connections. Additionally, when network utilization information ind=Lcates that a reduction in the rate of packets submittE~d onto the network is necessary, virtual ~1.644~9 connections with higher use of excess network resources incur higher packet submission rate reductions compared with virtual connections with lesser use of excess network resources.
In further accord with the present invention, bandwidth admis:ion for a virtual connection is controlled in accordance with assigned service categories for the determination of whether or not packets are admitted to the network, and for the identification of packets that arEa admitted to the network with an assigned priority for dei~ermin,ation of which packets will be discarded under certain congestion conditions.
In still further accord with the present invention, credits are associated with each service category such that various crE~dits ~3re assigned to a virtual connection per measurement period, T, for admission of packets to the network durp_ng the measurement period. Each packet is assigned a credit, and during extreme congestion conditions on the network, packets are discarded based on the priority of the corresponding assigned credit.
In further accord with the present invention, multiple classe~~ of scarvice may be designated which define the bandwidth admission, trunk queue and service discipline, and CUF measurement and SIR control for a given virtual connection.
The foregoing and other objects, features and advantages of tl~~e pre~~ent invention will become more apparent in light of i;he following detailed description X1.64489 of exemplary embodiments thereof as illustrated in the accompanying drawings.
Brief Descripti~~n of the Drawing's:
Fig. 1 is ;a schematic block diagram of a packet-based communications network;
Fig. 2 is ~~ schematic block diagram of an access module containers within a source node or a destination node of the communications network of Fig. 1;
Fig. 3 is ~~ schematic block diagram of a transit module contained within a node of the communications network of Fig. 1;
Fig. 4 is <3 diagram showing a packet having a header portion and a d<~ta portion;
Fig. 5 is ~~ grap:h depicting the measurement of one network round-trip delay (RTD) of a packet in the communications network; and Fig. 6 is ~~ graplh illustrating the relationship between rate cot~tro~.,and channel utilization factor (CUF) .
Best Mode for Carrying Out the Invention:
For purposE~s of simplifying the description of the present invention, number abbreviations are used in the following descrp_ption. The following table of abbreviations i~~ prov:ided as a reference for the reader:
~~6448~
ABBREV IATIOI'f= MEANING
ATM A:~ynchronous Transfer Mode CIR Committed Information Rate CMP Congestion Monitoring Period CPE Customer Premises Equipment CPU Central Processing Unit CUF Channel Utilization Factor EIR E;~cess Information Rate FF Fairness Factor ISR Initial Submission Rate PAD Packet Assembler/Disassembler RTD Round Trip Delay SIR Submission Information Rate VC Virtual Connection Referring t:o Fig. 1, a packet-based communication network 100 comprises a plurality of nodes including terminal or switching nodes 102 and intermediate nodes 105. As is welJ_ known in the art, in a packet-based communications network, packets are transmitted on a virtual connection from a source node 102a to a destination node 102b, usually traversing one or more intermediate nodes 10!5. The source node 102a and the destination nods' 102b are switching nodes 102 which interface with c:ustomE~r premises equipment (CPE) 108.
The customer premises equipment 108 may include a local area network, voice equipment, image and video equipment, or other data and packet transmission equipment which provides data in the form of packets for transmission via the packet-based communications network to a destination node.
ABBREV IATIOI'f= MEANING
ATM A:~ynchronous Transfer Mode CIR Committed Information Rate CMP Congestion Monitoring Period CPE Customer Premises Equipment CPU Central Processing Unit CUF Channel Utilization Factor EIR E;~cess Information Rate FF Fairness Factor ISR Initial Submission Rate PAD Packet Assembler/Disassembler RTD Round Trip Delay SIR Submission Information Rate VC Virtual Connection Referring t:o Fig. 1, a packet-based communication network 100 comprises a plurality of nodes including terminal or switching nodes 102 and intermediate nodes 105. As is welJ_ known in the art, in a packet-based communications network, packets are transmitted on a virtual connection from a source node 102a to a destination node 102b, usually traversing one or more intermediate nodes 10!5. The source node 102a and the destination nods' 102b are switching nodes 102 which interface with c:ustomE~r premises equipment (CPE) 108.
The customer premises equipment 108 may include a local area network, voice equipment, image and video equipment, or other data and packet transmission equipment which provides data in the form of packets for transmission via the packet-based communications network to a destination node.
z~~44~9 The customer premises equipment 108 is interconnected to a switching node 102 via an access module 110. The access module 110 interfaces with the customer premises equipment 108 and performs the following functions: packet segmentation and reassembly, bandwidth admission and enforcement, rate control, congestion recovery, generating alarms for congestion conditions, and generating accounting and statistics records. The switching node 102 also includes a transit module 112 which performs the following functions:
routing, signaling, congestion management, and measuring and reporting r~asource utilization.
An interconnection network 115 interconnects all the modules in a node. All of the nodal components can handle full-duplex communication. For example, an access module can concurrently switch packets from an access line 117 to the interconnection network 115, and vice-versa.
The intermEadiate nodes 105 include a plurality of transit modules 112 interconnected by an interconnection network 115. E<~ch of the transit modules 105 in an intermediate node 105 interface with a network backbone or trunkline 120 and perform all of the functions identified for i=he transit modules in the switching nodes 102. The trans_et modules 112 in the intermediate nodes 105 provide for the interconnection of the network trunklines 120 aria the interconnection networks 115 of the nodes 105.
Each virtu<~1 connection establishes a communications link between onE~ of tlhe network source nodes 102a and a w ~i6448~
corresponding network destination node 102b via intermediate nodes 105. An example of a virtual connection in Figure 1 is the connection between source node A and destination node D via intermediate nodes B
and C. Another virtual connection shown in Figure 1 is the virtual connection between source node A' and destination node D' via intermediate nodes B and C.
Although all of the virtual connections illustrated in Figure 1 utilized intermediate nodes 105, it will be understood by those skilled in the art that a virtual connection may be established directly between a source node 102a and a destination node 102b in accordance with the present invention.
As described hereinabove, all of the nodal components within the communications network 100 can handle full-duplex communications, and therefore, a switching node may be a source node in one virtual connection and ;~ destination node in another virtual connection, Fo:r example, in Figure 1, a virtual connection may lbe established between a source node A"
and destination node A' via intermediate nodes C and B.
During communications over a virtual connection, data generated ley customer premises equipment 108 is provided via the=_ access lines 117 to an access module 110 within a source node 102a. The data may be formatted in the appropriate packet format by the customer premises equipment 108, or alternatively, the access module 110 may include a known packet assembler/disassembler (PAD) for properly formatting data provided by the customer premises equipment 108 for provision to the communications network 100. Fig. 2 is a high level block ~~1_54489 diagram showing certain basic features of an access module 110 in a source node 102a. Referring to Fig. 2, the customer premises equipment 108 provides data to the access modules .110 on access lines 117. At least one dedicated virtual connection is established for each of the customer premises equipment 108 connected to the access module 110. Each of the virtual connections includes a queue 205, wherein packets provided by the customer premises equipment are initially stored, and a server 210, which controls the provision of the packets from the queue i~o the transit module 112. As described above, if the information or data provided by the customer premises equipment is not in proper packet format for provision to the communications network, a packet assemble~__°/disassembler 215 may be provided for formatting the data provided by the customer premises equipment 108 into the appropriate packet format. The various queues <ind servers associated with each of the virtual connections m;ay be implemented by a central processing unit (CPU) which, as is known in the art, includes appropriate processing and memory equipment for carrying out they functions described above and hereinafter with respect to the access module 110.
Referring now to Figs. 1 and 3, within the source node 102a, the »ntermediate nodes 105, and the destination node 102b, the packets associated with each virtual connectp_on are provided to and from the interconnection networks 115 by a transit module 112.
Each transit module 1:12 is arranged for full-duplex communication between the interconnection network 115 and a network trunkl_ine 1;?0. In each direction within the transit module x.12, a queue 305 is provided for receiving ~15448~
incoming packets from the interconnection network 115 or the network trunkline 120, and a server 310 is provided for retrieving :packets from the queue 305 and providing the packets to 'the network trunkline 120 or interconnection network 115, respectively.
In accordance with the present invention, when a virtual connection is established between a source node 102a and a destination node 102b, and possibly including one or more into=_rmediate nodes 105, packets initially flow along the virtual connection in a forward direction, i.e., from the :source node 102a to the destination node 102b. Within each node, the server 310 associated with the forward direction of the virtual connection takes a measurement of <~ channel utilization factor (CUF), which is a measure of the utilization of critical resources, e.g., bandwidth,, processing power, etc., during transmission of packets in a forward direction of the virtual connection. 'this utilization information is piggy-backed oni:o packets flowing in the backward (return) direction of the virtual connection by the server 310 associated with the return direction of the virtual path. =Cn the absence of return traffic, the destination nodE~ generates a special empty packet so that the utilization infarmation may be provided to the source node.
As describE~d in greater detail hereinafter, the CUF
information contained in the return packet traffic of a virtual connect~_on is updated during transmission such that when the rEaturn itraffic reaches the source node, the CUF information contained in the return traffic is indicative of the maximum degree of utilization indicated . ~1~4~89 by any of the resources within the virtual connection.
The CUF information provided in accordance with the present invention is provided to the source node in the most timely and efficient manner. The CUF information associated with the resource utilization experienced by a packet travelirn~ in the forward direction does not travel with that packet all the way to the destination node.
Instead, the current CUF information associated with a particular node is piggy-backed directly onto packets flowing in the :return direction to the source node. As will be underst~aod by those skilled in the art, in a virtual connection, the first node (access node) encountered by brackets traveling in the forward direction is typically th~a critical and potentially most limiting node (with respect to resource utilization) encountered by the packets in the virtual connection. The CUF
information related to the first node in the forward direction is the most timely information provided to the source node during packet transmission by the source node onto the virtual connection.
If the CUF indicates that the resources on the virtual connection are under-utilized, the source node instructs the v:irtual connection server 210 to increase its packet submission rate into the network. However, if the CUF indicatEas that the resources on the virtual connection are over-utilized, the source node instructs the virtual connection server 210 to reduce its packet submission rate into 'the network. The period of time between two consecutive rate adjustments is selected to allow the first adjustment to take effect throughout the network before t:he occurrence of the second adjustment.
This time is se7~.ected to be equal to at least one network ~1~4489 round-trip delay (RTD). A network round-trip delay is indicative of the time it takes for a packet transmitted from the source node to reach the destination node and return to the source node, via the virtual connection.
In a lightly loaded network, a typical RTD is approximately 80 to 100 milliseconds, and in a loaded network, a typical RTD is approximately 200 to 250 milliseconds. 'These times are used for example only and will vary based on the number of nodes, network l0 connections, etc.
The access modules 110 in the source node 102a controls the rate that packets are submitted to the network on a pe:r virtual connection basis. As described hereinabove, ea~~h virtual connection has a separate queue 205 and server :210, with the packet submission rate by the server 210 lbeing updated as a function of network feedback. The :rate at which packets are submitted to the network by a server 210 for a given virtual connection is 20 referred to as 'the submission information rate (SIR).
SIR is updated ~~s often as possible, but no faster than the network round-trip delay, except in extreme network congestion case: as described in greater detail hereinafter.
Referring ~~o Figs. 2 and 4, each packet 400 has a header portion ~~02 and a data portion 405. The header portion contain: various required network information such a routing, signaling, parity or error checking, and 30 various other information known to those skilled in the art. The data portion 405 contains the data (information) to be transmitted from the source node 102a to the destination node 102b. For measurement of the RTD
'1.64489 along each virtvual connection, each packet header portion 402 contains a wend flag (S) 410 and a receive flag (R) 411. A group of packets submitted to the network may be in the form of ;~ burst of packets, or the group may be a lengthy stream of packets. When the source node access module 110 updates SIR, it changes the value of the send flag 410, and places the new value of the send flag 410 into the header 402 of all packets 400 within the group being transmitted in the forward direction.
Additionally, the server 210 records the time that the first packet in the group of packets was transmitted onto the virtual connection. The time may be recorded in a memory register 220 of the server 210, for example. The server 210 in the destination node access module 110 reads the send :Flag 410 of all received packets 400 for a given virtual connection and places it into the receive flag of all pac)cets transmitted in the return direction.
The source node access module 110 waits until it detects that the receiv<~ flag 411 of packets received from the network for the given virtual connection is equal to the send flag 410 0~~ packets corresponding to the time stored in the memory re~giste:r 220. At this point, the module is assured that at least one round-trip period has elapsed since the last SIR update. RTD is measured as the time difference betwEaen the beginning of transmission of the group, i.e., them time stored in the memory register 220, and the time of receipt of the first return packet in the group.
The measurE~ment of round-trip delay is best understood by e3cample. Referring to Figs. 4 and 5, assume that a group o:f packets 400 provided by the source node access module 110 (Fig. 1) contains a send flag 410 ~;1~4~~9 equal to 1 (S = 1) and a receive flag 411 equal to 0 (R =
0). Therefore, each packet 400 within the group has a S
bit equal to 1 ~~nd an R bit equal to 0. When the first packet 400 within the group reaches the destination node access module 1:10 (Fig. 1), the value of the send flag 410 in each packet is read by the server 210 (Fig. 2) and placed into the receive flag 411 such that the receive flag of each packet is set equal to 1 (R = 1). Once the source node acccass module 110 (Fig. 1) detects that the receive flag 41:L of packets received is equal to the send flag 410 stored within the module, a round-trip delay is detected and a round-'trip delay time is determined as the time difference between the transmit time and the time that the return packet is received by the module.
The measurE~d RTD reflects the current network load (or delay) aloud the given virtual connection. Thus, SIR
updates occur more frequently under low network load conditions. A high frequency of SIR updates accelerates the frequency oi: end-to-end throughput increases for all active virtual connections. Under high network load conditions, howEaver, t3IR updates occur less frequently, thereby allowing the gate adjustments to propagate and take full effect. throughout the network. This feature is particularly beneficial for virtual connections which transmit bursty traffic. Bursty traffic creates short-lived, high network load when placed on the virtual connection, and the bursty traffic is likely to go off-line shortly thE~reafter. Less frequent SIR updates in this case allow the nE~twork to immediately transfer a large portion of the burst of traffic via the virtual connection before the virtual connections' SIR is throttled, as de~scribE~d hereinafter.
Referring to Fig. 1, the source node access module 110 of a virtual connection adjusts SIR as a function of the CUF received from the network. SIR is always greater than or equal t~~ a committed information rate (CIR), and less than or equal to an excess information rate (EIR).
CIR may be determined based on user subscription. For example, a user may subscribe to a CIR of 64 KBits/sec, or multiples thereof up to 512 KBits/sec, for example.
EIR may depend ~~n the type of access line used. For example, in the case of a Tl access line, EIR=1.536 MBits/sec. The value of SIR when customer premises equipment begins to provide packets for transmission via a virtual connection, or after a long period of virtual connection inactivity, is set equal to an initial submission rate (ISR). The relationship between CIR, ISR, and EIR is given below:
0 < CIR < ISR < EIR
ISR is a vcsry important parameter, especially for bursty traffic. ISR is determined per virtual connection, and its setting directly impacts the overall virtual connection delay performance. If ISR is set closer to CIR, i~hen the initial burst that a virtual connection submits into a network after a period of inactivity tend: to incur higher delay. This delay can be reduced significantly as ISR approaches EIR. However, a larger ISR sei~ting raises the possibility of short-term congestion at ccmvergence points in the network. This type of congestion may be alleviated by the combination of congestion management and end-to-end rate control.
ISR also has sic~nific,ant impact on the overall fairness of the rate coni~rol mechanism, as discussed hereinafter.
~'1644~9 During operation of a virtual connection, the source node access module 110 detects a round-trip period, and the module then attempts to match SIR to the available network resources along the virtual connection. Rate matching is accomplished by increasing or decreasing SIR
by a multiplicative factor, which is a function of the previous SIR, and the current CUF, CIR and EIR. If the current CUF indicates that the resources on the virtual connection are vunder-utilized, e.g., CUF is less than 100, the access module increases SIR. If the current value of CUF indicates that the resources on the virtual connection are ewer-utilized, e.g., CUF is greater than or equal to 100'x, the access module decreases SIR. The general behavior of SIR as a function of CUF is illustrated in :Figure 6.
After a long period of inactivity, the rate matching mechanism is go~,rerned by the following relationship for an idle virtual connection:
SIR = ISR
For an active virtual connection, SIR is given by the following relationship:
SIRi SIRj,I =MI MA CUF(a FF~),CIR ,EIR,(SIRjx Vii) Wherein FF; is a fairness factor, and a and /3 are constants selected to optimize network performance for specif is traf f ic: type;s . Typical values for a are between 0.95 and 1.0, and typical values for /3 are between 4.0 and 6Ø These values are provided for example only, and it will be understood by those skilled in the art that these values will vary based on the particular implementation of the invention.
SIR control is subject to the fairness factor (FFj to allocate res~curces fairly for all virtual connections, and to achieve quick SIR convergence during transient periods. FF is a measure of how close a virtual connections' SI1R is to EIR, such that FF is maximum when SIR = EIR and minimum when SIR = CIR, as given below:
SIRt-CIR
FFt= EIR-CIR X~
The parameter y is also selected to optimize network performance for specific traffic types. Typical values for ~y are in th<~ rage from 0 to 0.1.
FF has three maim effects on SIR. First, when a virtual connection becomes active, the combination of FF
and ISR for the newly active virtual connection forces other active vi~__°tual .connections to quickly reduce their use of excess rE~sourc~es in favor of the newly-active virtual connect_Lon. 'this allows a quick and fair allocation of rEasourc~as among active virtual connections.
Second, when a hate reduction is necessary, virtual connections with higher use of excess resources incur higher rate reduction compared to virtual connections with lesser use of excess resources. Third, when the network utilization changes, FF leads to a quick SIR
stabilization for all active virtual connections.
'~1E~44~9 CUF is the key factor of the traffic management system. It is 'the ratio of the demand for network resources to the total available resources. CUF is measured over a period of time, referred to as the congestion monitoring period (CMP). CMP is sufficiently long enough to ;allow an accurate statistical measurement of CUF, while a~t the same time, it is sufficiently short enough to allow several CUF measurements per RTD. The network resources being monitored are processor power and trunk bandwidth. The CUF value reported by each node is the maximum of both processor utilization and trunk utilization.
During a C1~IP, each transit module accumulates the total number of offered packets (both ingress and egress) for a given processor and the total number of offered bytes (egress only) for a given trunk. At the end of a CMP, the proces~~or utilization is calculated as the ratio of the total number of offered packets to the maximum switching capacity of the processor. The trunk utilization is calculated as the ratio of the total number of offered bytes to the maximum bandwidth capacity of the trunk. ;since CUF is a ratio of offered load to available capacity, C'UF can be greater than 1.
Referring i~o Fig. 4, CUF is encoded and placed in a CUF field 415 01_° the ;packet header 402. The CUF field 415, for examplca, may represent values ranging from 1~ to 1000$. In addii:ion, a special CUF value may be generated in response to Eaxtrem~e network congestion situations.
The special CUF value instructs the access module 110 (Fig. 1) of the virtual connection to immediately reduce SIR to CIR. Then special CUF may be indicated by a 2~~6 4489 special code placed in the CUF field 415. Alternatively, an additional field may be provided in the header portion 402 for the special code.
Referring ;gain to Fig. 1, the present invention provides a unique approach to communicating CUF values from network 100 to the source node 102a. Assume a virtual connection is established from source node A to destination node D via intermediate nodes B and C as illustrated in higure 1. As packets flow from nodes A to D, all transit modules 112 along the virtual connection measure processor utilization and trunk utilization. At the end of ever~~ CMP, each transit module inserts its measured CUF ini~o all packets flowing from nodes D to A
only if the mea:aured ~CUF is greater than the CUF stored in the packet. Therefore, packets that reach node A
contain the maximum CUF encountered along the virtual connections' forward path. The source node access module 110 uses this information to adjust the virtual connections' SIR accordingly.
As describEad hereinabove, the present invention does not communicate network utilization information to the destination nodE~, which in turn echoes the same information back; to the source node. This approach could cause the network utilization information to incur a full network round-trip delay before it reaches the source node. Instead, the present invention communicates utilization information directly to the source node, thereby all.owinc~ the utilization information to reach the source node quickly, with the worst-case delay being one half network round-trap delay.
Furthermore, the CUF communication mechanism of the present invention provides quick feedback, especially concerning the resource utilization of the most critical node in the network, typically the access node. The interconnection network of modules within a nude provides a high bandwidth medium which is generally not a bottleneck in inter-module communication. Hence, the access modules in node A have relatively unlimited access to the transit 'module. If the CPE traffic is bursty, the transit module :might become temporarily congested, hence critically impacting the network delay performance.
Therefore, the .access modules in node A require quick network feedback to adjust their SIRS in a timely fashion. The present invention allows node A to react quickly to internal congestion because the CUF field in backward packet, gets immediately updated by the transit module in the same node. In contrast, a rate control system that sends the congestion information from node A
to node D and then back to node A, would have the slowest reaction to congestion in the source node.
In addition to end-to-end rate control as described above, the present invention provides for queuing of packets during ~~eriods of network overload. Queuing in access modules :L10 is done on a per virtual connection basis. The serener 210 for each virtual connection queue 205 operates at a rate equal to the corresponding SIR.
Queuing in tran:~it modules is done on a per trunk basis.
Hence, all virtual connections established on a given trunk are queuecl< in a single queue. The server for this queue operates at the trunk rate.
~i6448~
In order to contain the end-to-end network delay and to detect incipient congestion, the occupancy of each queue is continuously monitored and compared to a set of queue threshold:a or congestion levels. For example, a transit module may monitor 9 congestion levels where level 0 corresponds to "no congestion" and level 8 corresponds to "extre:me congestion." Intermediate congestion leve:Ls are communicated to the source node access modules :L10 using the CUF mechanism described above. This causes tlhe access modules to reduce their SIR appropriate_Ly to .avoid congestion. If higher levels of congestion occur, the affected transit module might use the special CUF option to force all involved virtual connections to reduce their SIR to CIR. If all these congestion avoidance measures fail to prevent the network from reaching an extr~sme congestion situation, the network might bE~ ford=_d to selectively discard packets from the affectead queues as a congestion recovery measure as described hereinafter.
In addition to acing SIR for controlling the rate at which packets are admitted to the network as described hereinabove, a network access module 110 in a source node 102a may be responsible for admitting traffic to the network (bandwidth admission) according to assigned service subscription parameters. The service subscription parameters are associated with three categories of service including: committed, statistically committed, and excess.. The access module 110 uses a credit allocation scheme, based on the designated or assigned category of :service, to enforce bandwidth admission rules for admitting traffic to the network.
The committed ca~tegor~r of service (committed credits) ;~ls~4gg correspond to the minimum guaranteed bandwidth that is allocated for a given virtual connection. The statistically committed category of service (statistically committed credits) correspond to bandwidth that is statistically allocated for a given virtual connection. The excess category of service (excess credits) corres~~ond to any available bandwidth over and above committed and statistically committed bandwidth.
Credit usage per virtual connection is measured over an interval of mime, 'T, called a measurement interval.
Each virtual connection is allocated a certain number of credits, e.g., committed credits, statistically committed credits and excE~ss credits, to be used during each measurement int<srval.
As packets arrive at an access module 110 from customer premises equipment 108, the access module 110 performs bandwidth admission and enforcement every measurement intEarval 'T. The access module utilizes committed credii~s first, and marks the associated packets transmitted to i~he network with the committed credits as being committed packets. If the committed credits have been exhausted, the abscess module 110 utilizes the statisticaJ.ly committed credits, and marks the packets accordingly as statistically committed packets. Finally, if the statistically ~~ommitted credits have also been exhausted, the access module 110 utilizes the excess credits, and marks the associated packets as being excess packets. Once all credits have been exhausted during a measurement intE~rval, packets are discarded. Credits are refreshed at the beginning of every measurement interval T. These admission r,ales ensure that the network is not f looded with traf f is 'that violates the service subscription parameters associated with a give virtual connection.
The marking of packets allows the transit modules to selectively discard packets under extreme network congestion situations. For example, under extreme congestion situations, the transit modules will first discard excess packets. If the congestion persists, statistically committ~=d packets are discarded. Finally, under very severe congestions situations, committed packets are discarded. It will be understood by those skilled in the art that a congestion situation involving the discarding of committed packets is highly unlikely.
Although the invention is described as having three different service subscription parameters for bandwidth admission, it w»11 be understood by those skilled in the art that additional service subscription parameters may be defined for a given network application.
The present. invention may be utilized in a network having multiple classes of service. There are three network aspects that define a class of service:
1) bandwp.dth admission and enforcement at the network acc~=_ss point;
2) trunk queue and service discipline; and 3) CUF mE~asurement and its application to SIR
control.
21 g~ X89 An access module 110 might implement a different criteria of bandwidth admission, bandwidth enforcement, and SIR
control depending on the assigned class of service. For example, for a virtual connection associated with a high priority communications channel, a large number of committed credits may be assigned to ensure bandwidth admission. Additionally, a, ~3, and y are selected to ensure that SIR is responsive for communications of packets onto the network.
In one embodiment of the invention, each class of service might correspond to a different queue in the transit modules. In this case, a transit module 112 measures network resource utilization per queue, e.g., per virtual connection, and generates a CUF that applies to all virtual connecaions of the same class of service.
Alternatively, one queue may be provided for all classes of service. It. will therefore be understood that, depending on how transit queues are served, the CUF
measurement for a given class of service might also depend on the resource utilization of other classes of service that snare tree same trunk.
Although t:he invention has been described and illustrated with respect to exemplary embodiments thereof, it should bs: understood by those skilled in the art that the foregoing, and various other additions and omissions may be made, therein and thereto without departing from the spirit and scope of the present invention.
routing, signaling, congestion management, and measuring and reporting r~asource utilization.
An interconnection network 115 interconnects all the modules in a node. All of the nodal components can handle full-duplex communication. For example, an access module can concurrently switch packets from an access line 117 to the interconnection network 115, and vice-versa.
The intermEadiate nodes 105 include a plurality of transit modules 112 interconnected by an interconnection network 115. E<~ch of the transit modules 105 in an intermediate node 105 interface with a network backbone or trunkline 120 and perform all of the functions identified for i=he transit modules in the switching nodes 102. The trans_et modules 112 in the intermediate nodes 105 provide for the interconnection of the network trunklines 120 aria the interconnection networks 115 of the nodes 105.
Each virtu<~1 connection establishes a communications link between onE~ of tlhe network source nodes 102a and a w ~i6448~
corresponding network destination node 102b via intermediate nodes 105. An example of a virtual connection in Figure 1 is the connection between source node A and destination node D via intermediate nodes B
and C. Another virtual connection shown in Figure 1 is the virtual connection between source node A' and destination node D' via intermediate nodes B and C.
Although all of the virtual connections illustrated in Figure 1 utilized intermediate nodes 105, it will be understood by those skilled in the art that a virtual connection may be established directly between a source node 102a and a destination node 102b in accordance with the present invention.
As described hereinabove, all of the nodal components within the communications network 100 can handle full-duplex communications, and therefore, a switching node may be a source node in one virtual connection and ;~ destination node in another virtual connection, Fo:r example, in Figure 1, a virtual connection may lbe established between a source node A"
and destination node A' via intermediate nodes C and B.
During communications over a virtual connection, data generated ley customer premises equipment 108 is provided via the=_ access lines 117 to an access module 110 within a source node 102a. The data may be formatted in the appropriate packet format by the customer premises equipment 108, or alternatively, the access module 110 may include a known packet assembler/disassembler (PAD) for properly formatting data provided by the customer premises equipment 108 for provision to the communications network 100. Fig. 2 is a high level block ~~1_54489 diagram showing certain basic features of an access module 110 in a source node 102a. Referring to Fig. 2, the customer premises equipment 108 provides data to the access modules .110 on access lines 117. At least one dedicated virtual connection is established for each of the customer premises equipment 108 connected to the access module 110. Each of the virtual connections includes a queue 205, wherein packets provided by the customer premises equipment are initially stored, and a server 210, which controls the provision of the packets from the queue i~o the transit module 112. As described above, if the information or data provided by the customer premises equipment is not in proper packet format for provision to the communications network, a packet assemble~__°/disassembler 215 may be provided for formatting the data provided by the customer premises equipment 108 into the appropriate packet format. The various queues <ind servers associated with each of the virtual connections m;ay be implemented by a central processing unit (CPU) which, as is known in the art, includes appropriate processing and memory equipment for carrying out they functions described above and hereinafter with respect to the access module 110.
Referring now to Figs. 1 and 3, within the source node 102a, the »ntermediate nodes 105, and the destination node 102b, the packets associated with each virtual connectp_on are provided to and from the interconnection networks 115 by a transit module 112.
Each transit module 1:12 is arranged for full-duplex communication between the interconnection network 115 and a network trunkl_ine 1;?0. In each direction within the transit module x.12, a queue 305 is provided for receiving ~15448~
incoming packets from the interconnection network 115 or the network trunkline 120, and a server 310 is provided for retrieving :packets from the queue 305 and providing the packets to 'the network trunkline 120 or interconnection network 115, respectively.
In accordance with the present invention, when a virtual connection is established between a source node 102a and a destination node 102b, and possibly including one or more into=_rmediate nodes 105, packets initially flow along the virtual connection in a forward direction, i.e., from the :source node 102a to the destination node 102b. Within each node, the server 310 associated with the forward direction of the virtual connection takes a measurement of <~ channel utilization factor (CUF), which is a measure of the utilization of critical resources, e.g., bandwidth,, processing power, etc., during transmission of packets in a forward direction of the virtual connection. 'this utilization information is piggy-backed oni:o packets flowing in the backward (return) direction of the virtual connection by the server 310 associated with the return direction of the virtual path. =Cn the absence of return traffic, the destination nodE~ generates a special empty packet so that the utilization infarmation may be provided to the source node.
As describE~d in greater detail hereinafter, the CUF
information contained in the return packet traffic of a virtual connect~_on is updated during transmission such that when the rEaturn itraffic reaches the source node, the CUF information contained in the return traffic is indicative of the maximum degree of utilization indicated . ~1~4~89 by any of the resources within the virtual connection.
The CUF information provided in accordance with the present invention is provided to the source node in the most timely and efficient manner. The CUF information associated with the resource utilization experienced by a packet travelirn~ in the forward direction does not travel with that packet all the way to the destination node.
Instead, the current CUF information associated with a particular node is piggy-backed directly onto packets flowing in the :return direction to the source node. As will be underst~aod by those skilled in the art, in a virtual connection, the first node (access node) encountered by brackets traveling in the forward direction is typically th~a critical and potentially most limiting node (with respect to resource utilization) encountered by the packets in the virtual connection. The CUF
information related to the first node in the forward direction is the most timely information provided to the source node during packet transmission by the source node onto the virtual connection.
If the CUF indicates that the resources on the virtual connection are under-utilized, the source node instructs the v:irtual connection server 210 to increase its packet submission rate into the network. However, if the CUF indicatEas that the resources on the virtual connection are over-utilized, the source node instructs the virtual connection server 210 to reduce its packet submission rate into 'the network. The period of time between two consecutive rate adjustments is selected to allow the first adjustment to take effect throughout the network before t:he occurrence of the second adjustment.
This time is se7~.ected to be equal to at least one network ~1~4489 round-trip delay (RTD). A network round-trip delay is indicative of the time it takes for a packet transmitted from the source node to reach the destination node and return to the source node, via the virtual connection.
In a lightly loaded network, a typical RTD is approximately 80 to 100 milliseconds, and in a loaded network, a typical RTD is approximately 200 to 250 milliseconds. 'These times are used for example only and will vary based on the number of nodes, network l0 connections, etc.
The access modules 110 in the source node 102a controls the rate that packets are submitted to the network on a pe:r virtual connection basis. As described hereinabove, ea~~h virtual connection has a separate queue 205 and server :210, with the packet submission rate by the server 210 lbeing updated as a function of network feedback. The :rate at which packets are submitted to the network by a server 210 for a given virtual connection is 20 referred to as 'the submission information rate (SIR).
SIR is updated ~~s often as possible, but no faster than the network round-trip delay, except in extreme network congestion case: as described in greater detail hereinafter.
Referring ~~o Figs. 2 and 4, each packet 400 has a header portion ~~02 and a data portion 405. The header portion contain: various required network information such a routing, signaling, parity or error checking, and 30 various other information known to those skilled in the art. The data portion 405 contains the data (information) to be transmitted from the source node 102a to the destination node 102b. For measurement of the RTD
'1.64489 along each virtvual connection, each packet header portion 402 contains a wend flag (S) 410 and a receive flag (R) 411. A group of packets submitted to the network may be in the form of ;~ burst of packets, or the group may be a lengthy stream of packets. When the source node access module 110 updates SIR, it changes the value of the send flag 410, and places the new value of the send flag 410 into the header 402 of all packets 400 within the group being transmitted in the forward direction.
Additionally, the server 210 records the time that the first packet in the group of packets was transmitted onto the virtual connection. The time may be recorded in a memory register 220 of the server 210, for example. The server 210 in the destination node access module 110 reads the send :Flag 410 of all received packets 400 for a given virtual connection and places it into the receive flag of all pac)cets transmitted in the return direction.
The source node access module 110 waits until it detects that the receiv<~ flag 411 of packets received from the network for the given virtual connection is equal to the send flag 410 0~~ packets corresponding to the time stored in the memory re~giste:r 220. At this point, the module is assured that at least one round-trip period has elapsed since the last SIR update. RTD is measured as the time difference betwEaen the beginning of transmission of the group, i.e., them time stored in the memory register 220, and the time of receipt of the first return packet in the group.
The measurE~ment of round-trip delay is best understood by e3cample. Referring to Figs. 4 and 5, assume that a group o:f packets 400 provided by the source node access module 110 (Fig. 1) contains a send flag 410 ~;1~4~~9 equal to 1 (S = 1) and a receive flag 411 equal to 0 (R =
0). Therefore, each packet 400 within the group has a S
bit equal to 1 ~~nd an R bit equal to 0. When the first packet 400 within the group reaches the destination node access module 1:10 (Fig. 1), the value of the send flag 410 in each packet is read by the server 210 (Fig. 2) and placed into the receive flag 411 such that the receive flag of each packet is set equal to 1 (R = 1). Once the source node acccass module 110 (Fig. 1) detects that the receive flag 41:L of packets received is equal to the send flag 410 stored within the module, a round-trip delay is detected and a round-'trip delay time is determined as the time difference between the transmit time and the time that the return packet is received by the module.
The measurE~d RTD reflects the current network load (or delay) aloud the given virtual connection. Thus, SIR
updates occur more frequently under low network load conditions. A high frequency of SIR updates accelerates the frequency oi: end-to-end throughput increases for all active virtual connections. Under high network load conditions, howEaver, t3IR updates occur less frequently, thereby allowing the gate adjustments to propagate and take full effect. throughout the network. This feature is particularly beneficial for virtual connections which transmit bursty traffic. Bursty traffic creates short-lived, high network load when placed on the virtual connection, and the bursty traffic is likely to go off-line shortly thE~reafter. Less frequent SIR updates in this case allow the nE~twork to immediately transfer a large portion of the burst of traffic via the virtual connection before the virtual connections' SIR is throttled, as de~scribE~d hereinafter.
Referring to Fig. 1, the source node access module 110 of a virtual connection adjusts SIR as a function of the CUF received from the network. SIR is always greater than or equal t~~ a committed information rate (CIR), and less than or equal to an excess information rate (EIR).
CIR may be determined based on user subscription. For example, a user may subscribe to a CIR of 64 KBits/sec, or multiples thereof up to 512 KBits/sec, for example.
EIR may depend ~~n the type of access line used. For example, in the case of a Tl access line, EIR=1.536 MBits/sec. The value of SIR when customer premises equipment begins to provide packets for transmission via a virtual connection, or after a long period of virtual connection inactivity, is set equal to an initial submission rate (ISR). The relationship between CIR, ISR, and EIR is given below:
0 < CIR < ISR < EIR
ISR is a vcsry important parameter, especially for bursty traffic. ISR is determined per virtual connection, and its setting directly impacts the overall virtual connection delay performance. If ISR is set closer to CIR, i~hen the initial burst that a virtual connection submits into a network after a period of inactivity tend: to incur higher delay. This delay can be reduced significantly as ISR approaches EIR. However, a larger ISR sei~ting raises the possibility of short-term congestion at ccmvergence points in the network. This type of congestion may be alleviated by the combination of congestion management and end-to-end rate control.
ISR also has sic~nific,ant impact on the overall fairness of the rate coni~rol mechanism, as discussed hereinafter.
~'1644~9 During operation of a virtual connection, the source node access module 110 detects a round-trip period, and the module then attempts to match SIR to the available network resources along the virtual connection. Rate matching is accomplished by increasing or decreasing SIR
by a multiplicative factor, which is a function of the previous SIR, and the current CUF, CIR and EIR. If the current CUF indicates that the resources on the virtual connection are vunder-utilized, e.g., CUF is less than 100, the access module increases SIR. If the current value of CUF indicates that the resources on the virtual connection are ewer-utilized, e.g., CUF is greater than or equal to 100'x, the access module decreases SIR. The general behavior of SIR as a function of CUF is illustrated in :Figure 6.
After a long period of inactivity, the rate matching mechanism is go~,rerned by the following relationship for an idle virtual connection:
SIR = ISR
For an active virtual connection, SIR is given by the following relationship:
SIRi SIRj,I =MI MA CUF(a FF~),CIR ,EIR,(SIRjx Vii) Wherein FF; is a fairness factor, and a and /3 are constants selected to optimize network performance for specif is traf f ic: type;s . Typical values for a are between 0.95 and 1.0, and typical values for /3 are between 4.0 and 6Ø These values are provided for example only, and it will be understood by those skilled in the art that these values will vary based on the particular implementation of the invention.
SIR control is subject to the fairness factor (FFj to allocate res~curces fairly for all virtual connections, and to achieve quick SIR convergence during transient periods. FF is a measure of how close a virtual connections' SI1R is to EIR, such that FF is maximum when SIR = EIR and minimum when SIR = CIR, as given below:
SIRt-CIR
FFt= EIR-CIR X~
The parameter y is also selected to optimize network performance for specific traffic types. Typical values for ~y are in th<~ rage from 0 to 0.1.
FF has three maim effects on SIR. First, when a virtual connection becomes active, the combination of FF
and ISR for the newly active virtual connection forces other active vi~__°tual .connections to quickly reduce their use of excess rE~sourc~es in favor of the newly-active virtual connect_Lon. 'this allows a quick and fair allocation of rEasourc~as among active virtual connections.
Second, when a hate reduction is necessary, virtual connections with higher use of excess resources incur higher rate reduction compared to virtual connections with lesser use of excess resources. Third, when the network utilization changes, FF leads to a quick SIR
stabilization for all active virtual connections.
'~1E~44~9 CUF is the key factor of the traffic management system. It is 'the ratio of the demand for network resources to the total available resources. CUF is measured over a period of time, referred to as the congestion monitoring period (CMP). CMP is sufficiently long enough to ;allow an accurate statistical measurement of CUF, while a~t the same time, it is sufficiently short enough to allow several CUF measurements per RTD. The network resources being monitored are processor power and trunk bandwidth. The CUF value reported by each node is the maximum of both processor utilization and trunk utilization.
During a C1~IP, each transit module accumulates the total number of offered packets (both ingress and egress) for a given processor and the total number of offered bytes (egress only) for a given trunk. At the end of a CMP, the proces~~or utilization is calculated as the ratio of the total number of offered packets to the maximum switching capacity of the processor. The trunk utilization is calculated as the ratio of the total number of offered bytes to the maximum bandwidth capacity of the trunk. ;since CUF is a ratio of offered load to available capacity, C'UF can be greater than 1.
Referring i~o Fig. 4, CUF is encoded and placed in a CUF field 415 01_° the ;packet header 402. The CUF field 415, for examplca, may represent values ranging from 1~ to 1000$. In addii:ion, a special CUF value may be generated in response to Eaxtrem~e network congestion situations.
The special CUF value instructs the access module 110 (Fig. 1) of the virtual connection to immediately reduce SIR to CIR. Then special CUF may be indicated by a 2~~6 4489 special code placed in the CUF field 415. Alternatively, an additional field may be provided in the header portion 402 for the special code.
Referring ;gain to Fig. 1, the present invention provides a unique approach to communicating CUF values from network 100 to the source node 102a. Assume a virtual connection is established from source node A to destination node D via intermediate nodes B and C as illustrated in higure 1. As packets flow from nodes A to D, all transit modules 112 along the virtual connection measure processor utilization and trunk utilization. At the end of ever~~ CMP, each transit module inserts its measured CUF ini~o all packets flowing from nodes D to A
only if the mea:aured ~CUF is greater than the CUF stored in the packet. Therefore, packets that reach node A
contain the maximum CUF encountered along the virtual connections' forward path. The source node access module 110 uses this information to adjust the virtual connections' SIR accordingly.
As describEad hereinabove, the present invention does not communicate network utilization information to the destination nodE~, which in turn echoes the same information back; to the source node. This approach could cause the network utilization information to incur a full network round-trip delay before it reaches the source node. Instead, the present invention communicates utilization information directly to the source node, thereby all.owinc~ the utilization information to reach the source node quickly, with the worst-case delay being one half network round-trap delay.
Furthermore, the CUF communication mechanism of the present invention provides quick feedback, especially concerning the resource utilization of the most critical node in the network, typically the access node. The interconnection network of modules within a nude provides a high bandwidth medium which is generally not a bottleneck in inter-module communication. Hence, the access modules in node A have relatively unlimited access to the transit 'module. If the CPE traffic is bursty, the transit module :might become temporarily congested, hence critically impacting the network delay performance.
Therefore, the .access modules in node A require quick network feedback to adjust their SIRS in a timely fashion. The present invention allows node A to react quickly to internal congestion because the CUF field in backward packet, gets immediately updated by the transit module in the same node. In contrast, a rate control system that sends the congestion information from node A
to node D and then back to node A, would have the slowest reaction to congestion in the source node.
In addition to end-to-end rate control as described above, the present invention provides for queuing of packets during ~~eriods of network overload. Queuing in access modules :L10 is done on a per virtual connection basis. The serener 210 for each virtual connection queue 205 operates at a rate equal to the corresponding SIR.
Queuing in tran:~it modules is done on a per trunk basis.
Hence, all virtual connections established on a given trunk are queuecl< in a single queue. The server for this queue operates at the trunk rate.
~i6448~
In order to contain the end-to-end network delay and to detect incipient congestion, the occupancy of each queue is continuously monitored and compared to a set of queue threshold:a or congestion levels. For example, a transit module may monitor 9 congestion levels where level 0 corresponds to "no congestion" and level 8 corresponds to "extre:me congestion." Intermediate congestion leve:Ls are communicated to the source node access modules :L10 using the CUF mechanism described above. This causes tlhe access modules to reduce their SIR appropriate_Ly to .avoid congestion. If higher levels of congestion occur, the affected transit module might use the special CUF option to force all involved virtual connections to reduce their SIR to CIR. If all these congestion avoidance measures fail to prevent the network from reaching an extr~sme congestion situation, the network might bE~ ford=_d to selectively discard packets from the affectead queues as a congestion recovery measure as described hereinafter.
In addition to acing SIR for controlling the rate at which packets are admitted to the network as described hereinabove, a network access module 110 in a source node 102a may be responsible for admitting traffic to the network (bandwidth admission) according to assigned service subscription parameters. The service subscription parameters are associated with three categories of service including: committed, statistically committed, and excess.. The access module 110 uses a credit allocation scheme, based on the designated or assigned category of :service, to enforce bandwidth admission rules for admitting traffic to the network.
The committed ca~tegor~r of service (committed credits) ;~ls~4gg correspond to the minimum guaranteed bandwidth that is allocated for a given virtual connection. The statistically committed category of service (statistically committed credits) correspond to bandwidth that is statistically allocated for a given virtual connection. The excess category of service (excess credits) corres~~ond to any available bandwidth over and above committed and statistically committed bandwidth.
Credit usage per virtual connection is measured over an interval of mime, 'T, called a measurement interval.
Each virtual connection is allocated a certain number of credits, e.g., committed credits, statistically committed credits and excE~ss credits, to be used during each measurement int<srval.
As packets arrive at an access module 110 from customer premises equipment 108, the access module 110 performs bandwidth admission and enforcement every measurement intEarval 'T. The access module utilizes committed credii~s first, and marks the associated packets transmitted to i~he network with the committed credits as being committed packets. If the committed credits have been exhausted, the abscess module 110 utilizes the statisticaJ.ly committed credits, and marks the packets accordingly as statistically committed packets. Finally, if the statistically ~~ommitted credits have also been exhausted, the access module 110 utilizes the excess credits, and marks the associated packets as being excess packets. Once all credits have been exhausted during a measurement intE~rval, packets are discarded. Credits are refreshed at the beginning of every measurement interval T. These admission r,ales ensure that the network is not f looded with traf f is 'that violates the service subscription parameters associated with a give virtual connection.
The marking of packets allows the transit modules to selectively discard packets under extreme network congestion situations. For example, under extreme congestion situations, the transit modules will first discard excess packets. If the congestion persists, statistically committ~=d packets are discarded. Finally, under very severe congestions situations, committed packets are discarded. It will be understood by those skilled in the art that a congestion situation involving the discarding of committed packets is highly unlikely.
Although the invention is described as having three different service subscription parameters for bandwidth admission, it w»11 be understood by those skilled in the art that additional service subscription parameters may be defined for a given network application.
The present. invention may be utilized in a network having multiple classes of service. There are three network aspects that define a class of service:
1) bandwp.dth admission and enforcement at the network acc~=_ss point;
2) trunk queue and service discipline; and 3) CUF mE~asurement and its application to SIR
control.
21 g~ X89 An access module 110 might implement a different criteria of bandwidth admission, bandwidth enforcement, and SIR
control depending on the assigned class of service. For example, for a virtual connection associated with a high priority communications channel, a large number of committed credits may be assigned to ensure bandwidth admission. Additionally, a, ~3, and y are selected to ensure that SIR is responsive for communications of packets onto the network.
In one embodiment of the invention, each class of service might correspond to a different queue in the transit modules. In this case, a transit module 112 measures network resource utilization per queue, e.g., per virtual connection, and generates a CUF that applies to all virtual connecaions of the same class of service.
Alternatively, one queue may be provided for all classes of service. It. will therefore be understood that, depending on how transit queues are served, the CUF
measurement for a given class of service might also depend on the resource utilization of other classes of service that snare tree same trunk.
Although t:he invention has been described and illustrated with respect to exemplary embodiments thereof, it should bs: understood by those skilled in the art that the foregoing, and various other additions and omissions may be made, therein and thereto without departing from the spirit and scope of the present invention.
Claims (46)
1. A system for traffic management and congestion control on a virtual connection of a packet-based communications network, comprising:
customer premises equipment for providing packets for transmission on the virtual connection, and for receiving packets transmitted on the virtual connection;
a source node for accepting packets from corresponding source node customer premises equipment, and for transmitting the packets over the virtual connection through the packet-based communications network at a variable submission information rate (SIR);
a destination node for accepting the packets transmitted in a forward direction over the virtual connection from said source node to said destination node, for providing the packets transmitted in the forward direction to corresponding destination node customer premises equipment connected thereto, and for transmitting return packets in a return direction over the virtual connection from said destination node to said source node; and means for measuring a channel utilization factor (CUF), which is directly related to the degree of utilization of network resources, during transmission of packets in said forward direction, and for inserting said channel utilization factor in return packets travelling in said return direction;
wherein said source node comprises means for controlling said submission information rate in response to said channel utilization factor contained in said return packets.
customer premises equipment for providing packets for transmission on the virtual connection, and for receiving packets transmitted on the virtual connection;
a source node for accepting packets from corresponding source node customer premises equipment, and for transmitting the packets over the virtual connection through the packet-based communications network at a variable submission information rate (SIR);
a destination node for accepting the packets transmitted in a forward direction over the virtual connection from said source node to said destination node, for providing the packets transmitted in the forward direction to corresponding destination node customer premises equipment connected thereto, and for transmitting return packets in a return direction over the virtual connection from said destination node to said source node; and means for measuring a channel utilization factor (CUF), which is directly related to the degree of utilization of network resources, during transmission of packets in said forward direction, and for inserting said channel utilization factor in return packets travelling in said return direction;
wherein said source node comprises means for controlling said submission information rate in response to said channel utilization factor contained in said return packets.
2. A system according to claim 1, further comprising:
at least one intermediate node between said source node and said destination node in the virtual connection;
and wherein the channel utilization factor is measured at each node in the virtual connection; and wherein the channel utilization factor measured at a node is inserted into a return packet only if its magnitude is greater than the magnitude of the channel utilization factor previously inserted into the return packet.
at least one intermediate node between said source node and said destination node in the virtual connection;
and wherein the channel utilization factor is measured at each node in the virtual connection; and wherein the channel utilization factor measured at a node is inserted into a return packet only if its magnitude is greater than the magnitude of the channel utilization factor previously inserted into the return packet.
3. A system according to claim 2, wherein said source node and said destination node comprise:
at least one access module for interconnection with an access line connected to customer premises equipment, said access line providing a transmission path for packets between said access module and said customer premises equipment;
at least one transit module for interconnection with a network trunk line; and an interconnection network for interconnecting said access modules and said transit modules.
at least one access module for interconnection with an access line connected to customer premises equipment, said access line providing a transmission path for packets between said access module and said customer premises equipment;
at least one transit module for interconnection with a network trunk line; and an interconnection network for interconnecting said access modules and said transit modules.
4. A system according to claim 3, wherein each intermediate node comprises:
at least two transit modules, each for interconnection with a network trunk line; and an interconnection network for interconnecting said transit modules.
at least two transit modules, each for interconnection with a network trunk line; and an interconnection network for interconnecting said transit modules.
5. A system according to claim 4, wherein said access modules comprise:
at least one queue for queuing packets received from customer premises equipment for transmission on the virtual connection, and for queuing packets received from the virtual connection for provision to customer premises equipment, wherein at least one queue is provided for each customer premises equipment interconnected to said access module; and at least one server, responsive to said channel utilization factor contained in said return packets, for transmitting packets from said queue onto the virtual connection at said submission information rate, and for providing said return packets to said queue, at least one server being provided for each queue.
at least one queue for queuing packets received from customer premises equipment for transmission on the virtual connection, and for queuing packets received from the virtual connection for provision to customer premises equipment, wherein at least one queue is provided for each customer premises equipment interconnected to said access module; and at least one server, responsive to said channel utilization factor contained in said return packets, for transmitting packets from said queue onto the virtual connection at said submission information rate, and for providing said return packets to said queue, at least one server being provided for each queue.
6. A system according to claim 5, wherein said transit modules comprise:
an interconnection network queue for queuing packets received from the interconnection network;
a trunk line queue for queuing packets received from the trunk line;
an interconnection network server for transmitting packets from said interconnection network queue to the trunkline; and a trunkline server for transmitting packets from said trunkline queue to the interconnection network.
an interconnection network queue for queuing packets received from the interconnection network;
a trunk line queue for queuing packets received from the trunk line;
an interconnection network server for transmitting packets from said interconnection network queue to the trunkline; and a trunkline server for transmitting packets from said trunkline queue to the interconnection network.
7. A system according to claim 6, wherein said interconnection network server measures channel utilization factor as the greater of server utilization and trunkline utilization, wherein server utilization is the ratio of the total number of packets provided to said interconnection network queue or said trunkline queue to the maximum switching capacity of said interconnection network server or said trunkline server, respectively, and wherein trunkline utilization is the ratio of the total number of packets provided to the trunkline by said interconnection network server to the maximum bandwidth capacity of the trunkline.
8. A system according to claim 7, wherein said access module server updates the submission information rate once during each round-trip delay (RTD), wherein round-trip delay is the time it takes for a packet transmitted from the source node to reach the destination node and return to the source node over the virtual connection.
9. A system according to claim 8, wherein channel utilization factor is measured during a congestion monitoring period (CMP), and wherein said congestion monitoring period is sufficiently long enough to allow an accurate statistical measurement of the channel utilization factor and is sufficiently short enough to allow several measurements of channel utilization factor per round-trip delay.
10. A system according to claim 9, wherein:
submission information rate is always greater than or equal to a committed information rate (CIR);
submission information rate is always less than or equal to an excess information rate (EIR); and submission information rate is initially set equal to an initial submission rate (ISR), wherein the relationship between said committed information rate, said excess information rate, and said initial submission rate is given by:
CIR <= ISR <= EIR.
submission information rate is always greater than or equal to a committed information rate (CIR);
submission information rate is always less than or equal to an excess information rate (EIR); and submission information rate is initially set equal to an initial submission rate (ISR), wherein the relationship between said committed information rate, said excess information rate, and said initial submission rate is given by:
CIR <= ISR <= EIR.
11. A system according to claim 10, wherein after each round-trip delay, a new submission information rate (SIR i+1) is determined based on the current submission information rate (SIR i), said channel utilization factor, said committed information rate and said excess information rate, as given by the following relationship:
wherein FF i is a fairness factor, and .alpha. and .beta. are constants selected to optimize performance of the communications network.
wherein FF i is a fairness factor, and .alpha. and .beta. are constants selected to optimize performance of the communications network.
12. A system according to claim 11, wherein said fairness factor is given by the following relationship:
wherein .gamma. is a constant selected to optimize performance of the communications network.
wherein .gamma. is a constant selected to optimize performance of the communications network.
13. A system according to claim 12, wherein multiple classes of service are defined for the communications network, and wherein the magnitudes of said constants .alpha., .beta. and .gamma. are selected based on the class of service.
14. A system according to claim 10, further comprising means for providing a special channel utilization factor in response to extreme network congestion, said source node access modules reducing the submission information rate to the committed information rate in response to said special channel utilization factor.
15. A system according to claim 1, further comprising bandwidth admission means for admitting packets to the virtual connection according to assigned service subscription parameters.
16. A system according to claim 15, wherein said assigned service subscription parameters comprise:
committed credits corresponding to a minimum guaranteed bandwidth that is allocated on the communications network for the virtual connection;
statistically committed credits corresponding to bandwidth that is statistically allocated on the communications network for the virtual connection; and excess credits corresponding to any available bandwidth of the communications network that is greater than the statistically allocated and minimum guaranteed bandwidth.
committed credits corresponding to a minimum guaranteed bandwidth that is allocated on the communications network for the virtual connection;
statistically committed credits corresponding to bandwidth that is statistically allocated on the communications network for the virtual connection; and excess credits corresponding to any available bandwidth of the communications network that is greater than the statistically allocated and minimum guaranteed bandwidth.
17. A system according to claim 16, wherein said assigned service subscription parameters are allocated to the virtual connection on a per measurement interval basis, wherein the number of committed credits, statistically committed credits and excess credits allocated to the virtual connection are updated at the end of each measurement interval, and wherein during a measurement interval, said source node first provides a number of packets to the virtual connection corresponding to the number of committed credits, and marks such packets as committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of statistically committed credits, and marks such packets as statistically committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of excess credits, and marks such packets as excess packets;
and said source node discards any remaining packets.
said source node next provides a number of packets to the virtual connection corresponding to the number of statistically committed credits, and marks such packets as statistically committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of excess credits, and marks such packets as excess packets;
and said source node discards any remaining packets.
18. A system according to claim 17, wherein during transmission on the virtual connection, packets are selectively discarded under certain network congestion situations according to a predefined protocol wherein excess packets are discarded first, statistically committed packets are discarded second, and committed packets are discarded last.
19. A system according to claim 1, wherein said source node and said destination node each comprise:
at least one access module for interconnection with an access line connected to customer premises equipment, said access line providing a transmission path for packets between said access module and said customer premises equipment;
at least one transit module for interconnection with a network trunk line; and an interconnection network for interconnecting said access modules and said transit modules.
at least one access module for interconnection with an access line connected to customer premises equipment, said access line providing a transmission path for packets between said access module and said customer premises equipment;
at least one transit module for interconnection with a network trunk line; and an interconnection network for interconnecting said access modules and said transit modules.
20. A system according to claim 19, wherein the channel utilization factor is measured at each transit module in the virtual connection, and wherein the channel utilization factor measured at a transit module is inserted into a return packet only if its magnitude is greater than the magnitude of the channel utilization factor previously inserted into the return packet.
21. A system according to claim 20, wherein said access modules comprise:
at least one queue for queuing packets received from customer premises equipment for transmission on the virtual connection, and for queuing packets received from the virtual connection for provision to customer premises equipment, wherein at least one queue is provided for each customer premises equipment interconnected to said access module; and at least one server, responsive to said channel utilization factor contained in said return packets, for transmitting packets from said queue onto the virtual connection at said submission information rate, and for providing said return packets to said queue, at least one server being provided for each queue.
at least one queue for queuing packets received from customer premises equipment for transmission on the virtual connection, and for queuing packets received from the virtual connection for provision to customer premises equipment, wherein at least one queue is provided for each customer premises equipment interconnected to said access module; and at least one server, responsive to said channel utilization factor contained in said return packets, for transmitting packets from said queue onto the virtual connection at said submission information rate, and for providing said return packets to said queue, at least one server being provided for each queue.
22. A system according to claim 21, wherein said transit modules comprise:
an interconnection network queue for queuing packets received from the interconnection network;
a trunk line queue for queuing packets received from the trunk line;
an interconnection network server for transmitting packets from said interconnection network queue to the trunkline; and a trunkline server for transmitting packets from said trunkline queued to the interconnection network.
an interconnection network queue for queuing packets received from the interconnection network;
a trunk line queue for queuing packets received from the trunk line;
an interconnection network server for transmitting packets from said interconnection network queue to the trunkline; and a trunkline server for transmitting packets from said trunkline queued to the interconnection network.
23. A system according to claim 22, wherein said interconnection network server measures channel utilization factor as the greater of server utilization and trunkline utilization, wherein server utilization is the ratio of the total number of packets provided to said interconnection network queue or said trunkline queue to the maximum switching capacity of said interconnection network server or said trunkline server, respectively, and wherein trunkline utilization is the ratio of the total number of packets provided to the trunkline by said interconnection network server to the maximum bandwidth capacity of the trunkline.
24. A system according to claim 23, wherein said access module server updates the submission information rate once during each round-trip delay (RTD), wherein round-trip delay is the time it takes for a packet transmitted from the source node to reach the destination node and return to the source node over the virtual connection.
25. A system according to claim 24, wherein channel utilization factor is measured during a congestion monitoring period (CMP), and wherein said congestion monitoring period is sufficiently long to allow an accurate statistical measurement of the channel utilization factor and is sufficiently short enough to allow several measurements of channel utilization factor per round-trip delay.
26. A system according to claim 25, wherein:
submission information rate is always greater than or equal to a committed information rate (CIR);
submission information rate is always less than or equal to an excess information rate (EIR); and submission information rate is initially set equal to an initial submission rate (ISR), wherein the relationship between said committed information rate, said excess information rate, and said initial submission rate is given by:
CIR <= ISR <= EIR.
submission information rate is always greater than or equal to a committed information rate (CIR);
submission information rate is always less than or equal to an excess information rate (EIR); and submission information rate is initially set equal to an initial submission rate (ISR), wherein the relationship between said committed information rate, said excess information rate, and said initial submission rate is given by:
CIR <= ISR <= EIR.
27. A system according to claim 26, further comprising means for providing a special channel utilization factor in response to extreme network congestion, said source node access modules reducing the submission information rate to the committed information rate in response to said special channel utilization factor.
28. A system according to claim 27, wherein a round-trip delay is determined as the time it takes for a packet to travel from said source node to said destination node and back over the virtual connection, and after each round-trip delay, a new submission information rate (SIR i+1) is determined based on the current submission information rate (SIR i), said connection utilization factor, said committed information rate and said excess information rate, as given by the following relationship:
wherein FF i is a fairness factor, and .alpha. and .beta. are constants selected to optimize performance of the communications network.
wherein FF i is a fairness factor, and .alpha. and .beta. are constants selected to optimize performance of the communications network.
29. A system according to claim 28, wherein said fairness factor is given by the following relationship:
wherein .gamma. is a constant selected to optimize performance of the communications network.
wherein .gamma. is a constant selected to optimize performance of the communications network.
30. A system according to claim 28, wherein multiple classes of service are defined for the communications network, and wherein the magnitudes of said constants .alpha., .beta. and .gamma. are selected based on the class of service.
31. A system according to claim 1, wherein:
submission information rate is always greater than or equal to a committed information rate (CIR);
submission information rate is always less than or equal to an excess information rate (EIR); and submission information rate is initially set equal to an initial submission rate (ISR), wherein the relationship between said committed information rate, said excess information rate, and said initial submission rate is given by:
CIR <= ISR <= EIR.
submission information rate is always greater than or equal to a committed information rate (CIR);
submission information rate is always less than or equal to an excess information rate (EIR); and submission information rate is initially set equal to an initial submission rate (ISR), wherein the relationship between said committed information rate, said excess information rate, and said initial submission rate is given by:
CIR <= ISR <= EIR.
32. A system according to claim 31, wherein after each round-trip delay, a new submission information rate (SIR i+1) is determined based on the current submission information rate (SIR i), said connection utilization factor, said committed information rate and said excess information rate, as given by the following relationship:
wherein FF i is a fairness factor, and .alpha. and .beta. are constants selected to optimize performance of the communications network.
wherein FF i is a fairness factor, and .alpha. and .beta. are constants selected to optimize performance of the communications network.
33. A system according to claim 32, wherein said fairness factor is given by the following relationship:
wherein .gamma. is a constant selected to optimize performance of the communications network.
wherein .gamma. is a constant selected to optimize performance of the communications network.
34. A system according to claim 33, further comprising bandwidth admission means for admitting packets to the virtual connection according to assigned service subscription parameters comprising:
committed credits corresponding to a minimum guaranteed bandwidth that is allocated on the communications network for the virtual connection;
statistically committed credits corresponding to bandwidth that is statistically allocated on the communications network for the virtual connection; and excess credits corresponding to any available bandwidth of the communications network that is over and above the statistically allocated and minimum guaranteed bandwidth.
committed credits corresponding to a minimum guaranteed bandwidth that is allocated on the communications network for the virtual connection;
statistically committed credits corresponding to bandwidth that is statistically allocated on the communications network for the virtual connection; and excess credits corresponding to any available bandwidth of the communications network that is over and above the statistically allocated and minimum guaranteed bandwidth.
35. A system according to claim 34, wherein said assigned service subscription parameters are allocated to the virtual connection on a per measurement interval basis, wherein the number of committed credits, statistically committed credits and excess credits allocated to the virtual connection is updated at the end of each measurement interval, and wherein during a measurement interval, said source node first provides a number of packets to the virtual connection corresponding to the number of committed credits, and marks such packets as committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of statistically committed credits, and marks such packets as statistically committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of excess credits, and marks such packets as excess packets;
and said source node discards any remaining packets.
said source node next provides a number of packets to the virtual connection corresponding to the number of statistically committed credits, and marks such packets as statistically committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of excess credits, and marks such packets as excess packets;
and said source node discards any remaining packets.
36. A system according to claim 35, wherein during transmission on the virtual connection, packets are selectively discarded under certain network congestion situations according to a predefined protocol wherein excess packets are discarded first, statistically committed packets are discarded second, and committed packets are discarded last.
37. A system according to claim 36, wherein multiple classes of service are defined for the communications network, and wherein the magnitudes of said constants .alpha., .beta. and .gamma. are, and the amount of assigned committed credits, statistically committed credits and excess credits are selected based on the class of service.
38. A method for traffic management and congestion control of packets traveling in a forward direction from a source node to a destination node on a virtual connection of a packet-based communications network, the method comprising the steps of:
transmitting the packets from the source node to the destination node on the virtual connection at a variable submission information rate (SIR);
measuring a channel utilization factor (CUF), which is directly related to the degree of utilization of network resources, during transmission of packets in said forward direction;
inserting said channel utilization factor in return packets travelling in a return direction from the destination node to the source node on the virtual connection; and adjusting said submission information rate in response to said channel utilization factor contained in said return packets.
transmitting the packets from the source node to the destination node on the virtual connection at a variable submission information rate (SIR);
measuring a channel utilization factor (CUF), which is directly related to the degree of utilization of network resources, during transmission of packets in said forward direction;
inserting said channel utilization factor in return packets travelling in a return direction from the destination node to the source node on the virtual connection; and adjusting said submission information rate in response to said channel utilization factor contained in said return packets.
39. A method according to claim 38, wherein:
submission information rate is always greater than or equal to a committed information rate (CIR);
submission information rate is always less than or equal to an excess information rate (EIR); and submission information rate is initially set equal to an initial submission rate (ISR), wherein the relationship between said committed information rate, said excess information rate, and said initial submission rate is given by:
CIR <= ISR <= EIR.
submission information rate is always greater than or equal to a committed information rate (CIR);
submission information rate is always less than or equal to an excess information rate (EIR); and submission information rate is initially set equal to an initial submission rate (ISR), wherein the relationship between said committed information rate, said excess information rate, and said initial submission rate is given by:
CIR <= ISR <= EIR.
40. A method according to claim 39, further comprising the steps of:
measuring a round-trip delay as the time it take a packet to travel from the source node to the destination node and back over the virtual connection; and determining a new submission information rate (SIR i+1) after each round-trip delay based on the current submission information rate (SIR i), said connection utilization factor, said committed information rate and said excess information rate, as given by the following relationship:
wherein FF i is a fairness factor, and .alpha. and .beta. are constants selected to optimize performance of the communications network.
measuring a round-trip delay as the time it take a packet to travel from the source node to the destination node and back over the virtual connection; and determining a new submission information rate (SIR i+1) after each round-trip delay based on the current submission information rate (SIR i), said connection utilization factor, said committed information rate and said excess information rate, as given by the following relationship:
wherein FF i is a fairness factor, and .alpha. and .beta. are constants selected to optimize performance of the communications network.
41. A method according to claim 40, wherein said fairness factor is given by the following relationship:
wherein .gamma. is a constant selected to optimize performance of the communications network.
wherein .gamma. is a constant selected to optimize performance of the communications network.
42. A method according to claim 41, further comprising the step of admitting packets to the virtual connection according to assigned service subscription parameters.
43. A method according to claim 42, wherein said assigned service subscription parameters comprise:
committed credits corresponding to a minimum guaranteed bandwidth that is allocated on the communications network for the virtual connection;
statistically committed credits corresponding to bandwidth that is statistically allocated on the communications network for the virtual connection; and excess credits, corresponding to any available bandwidth of the communications network that is over and above the statistically allocated and minimum guaranteed bandwidth.
committed credits corresponding to a minimum guaranteed bandwidth that is allocated on the communications network for the virtual connection;
statistically committed credits corresponding to bandwidth that is statistically allocated on the communications network for the virtual connection; and excess credits, corresponding to any available bandwidth of the communications network that is over and above the statistically allocated and minimum guaranteed bandwidth.
44. A system according to claim 43, further comprising the steps of:
allocating said assigned service subscription parameters to the virtual connection on a per measurement interval basis;
updating the number of committed credits, statistically committed credits and excess credits allocated to the virtual connection at the end of each measurement interval, and during a measurement interval:
said source node first provides a number of packets to the virtual connection corresponding to the number of committed credits, and marks such packets as committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of statistically committed credits, and marks such packets as statistically committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of excess credits, and marks such packets as excess packets;
and said source node discards any remaining packets.
allocating said assigned service subscription parameters to the virtual connection on a per measurement interval basis;
updating the number of committed credits, statistically committed credits and excess credits allocated to the virtual connection at the end of each measurement interval, and during a measurement interval:
said source node first provides a number of packets to the virtual connection corresponding to the number of committed credits, and marks such packets as committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of statistically committed credits, and marks such packets as statistically committed packets;
said source node next provides a number of packets to the virtual connection corresponding to the number of excess credits, and marks such packets as excess packets;
and said source node discards any remaining packets.
45. A method according to claim 44, further comprising the step of selectively discarding packets from a virtual connection under certain network congestion situations according to a predefined protocol wherein excess packets are discarded first, statistically committed packets are discarded second, and committed packets are discarded last.
46. A method according to claim 45, further comprising the step of providing a special channel utilization factor in response to extreme network congestion, said source node access modules reducing the submission information rate to the committed information rate in response to said special channel utilization factor.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US35874194A | 1994-12-19 | 1994-12-19 | |
US08/358,741 | 1994-12-19 |
Publications (1)
Publication Number | Publication Date |
---|---|
CA2164489A1 true CA2164489A1 (en) | 1996-06-20 |
Family
ID=23410853
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA002164489A Abandoned CA2164489A1 (en) | 1994-12-19 | 1995-12-05 | Traffic management and congestion control for packet-based networks |
Country Status (13)
Country | Link |
---|---|
US (1) | US5633861A (en) |
EP (1) | EP0719012B1 (en) |
JP (1) | JP3833739B2 (en) |
KR (1) | KR960027839A (en) |
CN (1) | CN1111999C (en) |
AR (1) | AR000331A1 (en) |
AT (1) | ATE218259T1 (en) |
AU (1) | AU703410B2 (en) |
BR (1) | BR9505887A (en) |
CA (1) | CA2164489A1 (en) |
DE (1) | DE69526839T2 (en) |
ES (1) | ES2177600T3 (en) |
ZA (1) | ZA959722B (en) |
Families Citing this family (146)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6847611B1 (en) | 1990-12-10 | 2005-01-25 | At&T Corp. | Traffic management for frame relay switched data service |
EP0712220A1 (en) * | 1994-11-08 | 1996-05-15 | International Business Machines Corporation | Hop-by-hop flow control in an ATM network |
JPH08139737A (en) * | 1994-11-14 | 1996-05-31 | Nec Corp | Congestion control system |
JP2921424B2 (en) * | 1995-01-13 | 1999-07-19 | 日本電気株式会社 | ATM electronic exchange network system and electronic exchange used in the system |
DE69636242T8 (en) * | 1995-02-23 | 2007-10-04 | Ntt Docomo, Inc. | METHOD AND DEVICE FOR MULTI-CELL TRANSMISSION |
KR19980702938A (en) | 1995-03-08 | 1998-09-05 | 피들러 피터 | Broadband switching system |
WO1996027966A1 (en) * | 1995-03-08 | 1996-09-12 | Oxford Brookes University | Broadband switching system |
DE69638195D1 (en) | 1995-03-08 | 2010-07-22 | British Telecomm | BROADBAND LEARNING NETWORK |
GB9511319D0 (en) * | 1995-06-05 | 1995-08-02 | Gen Datacomm Adv Res | Controlling the flow of ATM cells in an ATM network |
JPH098814A (en) * | 1995-06-21 | 1997-01-10 | Fuji Xerox Co Ltd | Data transmission device and method thereof and data communication system |
GB9514206D0 (en) | 1995-07-12 | 1995-09-13 | British Telecomm | Rate controller |
US5996019A (en) | 1995-07-19 | 1999-11-30 | Fujitsu Network Communications, Inc. | Network link access scheduling using a plurality of prioritized lists containing queue identifiers |
US5828846A (en) * | 1995-11-22 | 1998-10-27 | Raptor Systems, Inc. | Controlling passage of packets or messages via a virtual connection or flow |
US5815667A (en) * | 1995-11-28 | 1998-09-29 | Ncr Corporation | Circuits and methods for intelligent acknowledgement based flow control in a processing system network |
WO1997025830A1 (en) | 1996-01-09 | 1997-07-17 | British Telecommunications Public Limited Company | A service multiplexer |
US5809013A (en) * | 1996-02-09 | 1998-09-15 | Interactive Technologies, Inc. | Message packet management in a wireless security system |
US5777984A (en) * | 1996-04-01 | 1998-07-07 | Motorola Inc. | Method and apparatus for controlling cell transmission rate in a cell based network in the presence of congestion |
US5812527A (en) * | 1996-04-01 | 1998-09-22 | Motorola Inc. | Simplified calculation of cell transmission rates in a cell based netwook |
US6973034B1 (en) * | 1999-06-29 | 2005-12-06 | Cisco Technology, Inc. | Technique for collecting operating information from network elements, and for controlling network element behavior in a feedback-based, adaptive data network |
US6034945A (en) | 1996-05-15 | 2000-03-07 | Cisco Technology, Inc. | Method and apparatus for per traffic flow buffer management |
US5878224A (en) * | 1996-05-24 | 1999-03-02 | Bell Communications Research, Inc. | System for preventing server overload by adaptively modifying gap interval that is used by source to limit number of transactions transmitted by source to server |
FR2750283B1 (en) * | 1996-06-20 | 1998-07-31 | Quinquis Jean Paul | LOCAL MOBILE ACCESS NETWORK PROVIDED WITH MEANS FOR MANAGING RESOURCES IN SUCH A NETWORK |
US6961341B1 (en) * | 1996-07-02 | 2005-11-01 | Microsoft Corporation | Adaptive bandwidth throttling for network services |
US5799002A (en) * | 1996-07-02 | 1998-08-25 | Microsoft Corporation | Adaptive bandwidth throttling for network services |
FR2754123B1 (en) * | 1996-10-01 | 1999-01-08 | Alsthom Cge Alcatel | TRAFFIC MANAGEMENT AND CONGESTION CONTROL SYSTEM FOR PACKET-BASED NETWORKS |
SE507754C2 (en) * | 1996-10-10 | 1998-07-13 | Ericsson Telefon Ab L M | Method and device for network configuration at high load |
US5901138A (en) * | 1996-11-08 | 1999-05-04 | International Business Machines Corporation | Process for improving the efficiency of high performance routing |
US5951644A (en) * | 1996-12-24 | 1999-09-14 | Apple Computer, Inc. | System for predicting and managing network performance by managing and monitoring resourse utilization and connection of network |
JP3193947B2 (en) * | 1997-01-08 | 2001-07-30 | 株式会社ディジタル・ビジョン・ラボラトリーズ | Data transmission system and data transmission method |
SE9700714L (en) * | 1997-02-27 | 1998-08-28 | Ericsson Telefon Ab L M | Method of controlling the data packet flow in a switch unit and a switch unit adapted to operate according to the method |
SE519482C2 (en) * | 1997-03-07 | 2003-03-04 | Telia Ab | Method and apparatus of a telecommunications system for determining the amount of necessary resources in a given traffic situation |
SE510170C2 (en) | 1997-05-16 | 1999-04-26 | Ericsson Telefon Ab L M | Method and apparatus for controlling data packet flow by back pressure message |
JPH10336202A (en) * | 1997-06-03 | 1998-12-18 | Fuji Xerox Co Ltd | Data transfer device and its method |
US7260060B1 (en) | 1997-06-07 | 2007-08-21 | Nortel Networks Limited | Call admission control |
US6798786B1 (en) * | 1999-06-07 | 2004-09-28 | Nortel Networks Limited | Managing calls over a data network |
US6243751B1 (en) * | 1997-06-11 | 2001-06-05 | Oracle Corporation | Method and apparatus for coupling clients to servers |
US6088728A (en) * | 1997-06-11 | 2000-07-11 | Oracle Corporation | System using session data stored in session data storage for associating and disassociating user identifiers for switching client sessions in a server |
US6430191B1 (en) | 1997-06-30 | 2002-08-06 | Cisco Technology, Inc. | Multi-stage queuing discipline |
US6487202B1 (en) | 1997-06-30 | 2002-11-26 | Cisco Technology, Inc. | Method and apparatus for maximizing memory throughput |
US6201813B1 (en) * | 1997-06-30 | 2001-03-13 | Cisco Technology, Inc. | Method and apparatus for using ATM queues for segmentation and reassembly of data frames |
US6188671B1 (en) * | 1997-07-03 | 2001-02-13 | At&T Corp | Traffic management for frame relay switched data service |
US6081524A (en) | 1997-07-03 | 2000-06-27 | At&T Corp. | Frame relay switched data service |
DE19729590A1 (en) | 1997-07-10 | 1999-01-14 | Siemens Ag | Procedure for optimizing the utilization on link sections in ABR traffic |
US5995487A (en) * | 1997-08-06 | 1999-11-30 | Northern Telecom Limited | Method and apparatus for providing subscribers with unguaranteed bandwidth connections wherein subscribers are alerted by encoded messages or a warning tone |
US6343085B1 (en) | 1997-08-28 | 2002-01-29 | Microsoft Corporation | Adaptive bandwidth throttling for individual virtual services supported on a network server |
GB2331659A (en) * | 1997-11-21 | 1999-05-26 | Ericsson Telefon Ab L M | Resource reservation |
CA2240596A1 (en) * | 1997-11-28 | 1999-05-28 | Newbridge Networks Corporation | Controlling atm layer transfer characteristics based on physical layer dynamic rate adaptation |
US6526060B1 (en) | 1997-12-05 | 2003-02-25 | Cisco Technology, Inc. | Dynamic rate-based, weighted fair scheduler with explicit rate feedback option |
AU1710099A (en) | 1997-12-08 | 1999-06-28 | Packeteer, Inc. | Method for data rate control for heterogenous or peer internetworking |
KR100257989B1 (en) * | 1997-12-09 | 2000-06-01 | 고대식 | Network simulation method |
US6628616B2 (en) * | 1998-01-30 | 2003-09-30 | Alcatel | Frame relay network featuring frame relay nodes with controlled oversubscribed bandwidth trunks |
CA2266307A1 (en) * | 1998-03-24 | 1999-09-24 | Northern Telecom Limited | Apparatus and method for optimizing max-min fair rate control in abr sessions |
US6597662B1 (en) | 1998-03-24 | 2003-07-22 | Nortel Networks Limited | Apparatus and method for optimizing max-min fair rate control in ABR sessions |
WO1999052299A2 (en) * | 1998-04-03 | 1999-10-14 | Netcore Systems, Inc. | High speed router-switch |
JP2955561B1 (en) | 1998-05-29 | 1999-10-04 | 株式会社ディジタル・ビジョン・ラボラトリーズ | Stream communication system and stream transfer control method |
US6097697A (en) * | 1998-07-17 | 2000-08-01 | Sitara Networks, Inc. | Congestion control |
US6434559B1 (en) * | 1998-10-09 | 2002-08-13 | Xpandable Technology, Inc. | Critical resource management |
US6370579B1 (en) * | 1998-10-21 | 2002-04-09 | Genuity Inc. | Method and apparatus for striping packets over parallel communication links |
US9239763B2 (en) | 2012-09-28 | 2016-01-19 | Oracle International Corporation | Container database |
US6504818B1 (en) * | 1998-12-03 | 2003-01-07 | At&T Corp. | Fair share egress queuing scheme for data networks |
US6711137B1 (en) | 1999-03-12 | 2004-03-23 | International Business Machines Corporation | System and method for analyzing and tuning a communications network |
US6768717B1 (en) | 1999-05-28 | 2004-07-27 | Network Equipment Technologies, Inc. | Apparatus and method for traffic shaping in a network switch |
US6628610B1 (en) * | 1999-06-28 | 2003-09-30 | Cisco Technology, Inc. | Methods and apparatus for managing a flow of packets using change and reply signals |
US6765864B1 (en) | 1999-06-29 | 2004-07-20 | Cisco Technology, Inc. | Technique for providing dynamic modification of application specific policies in a feedback-based, adaptive data network |
US6570851B1 (en) * | 1999-07-01 | 2003-05-27 | Nokia Telecommunications Oy | Receiver driven differentiated service marking for unicast and multicast applications |
US7016301B1 (en) * | 1999-09-01 | 2006-03-21 | Cisco Technology, Inc. | Fair multiplexing scheme for multiple input port router |
US6952396B1 (en) | 1999-09-27 | 2005-10-04 | Nortel Networks Limited | Enhanced dual counter rotating ring network control system |
JP3391316B2 (en) | 1999-10-22 | 2003-03-31 | 日本電気株式会社 | Network system |
US6539340B1 (en) | 1999-12-10 | 2003-03-25 | Cisco Technology, Inc. | Methods and apparatus for measuring resource usage within a computer system |
US6587438B1 (en) * | 1999-12-22 | 2003-07-01 | Resonate Inc. | World-wide-web server that finds optimal path by sending multiple syn+ack packets to a single client |
US6751198B1 (en) * | 1999-12-27 | 2004-06-15 | Nortel Networks Limited | System and method for measuring round trip delay of voice packets in a telephone system |
US6775292B1 (en) | 2000-01-24 | 2004-08-10 | Cisco Technology, Inc. | Method for servicing of multiple queues carrying voice over virtual circuits based on history |
EP1256210A2 (en) * | 2000-02-04 | 2002-11-13 | HRL Laboratories, LLC | System and method for pricing-based quality of service |
US6850530B1 (en) | 2000-02-04 | 2005-02-01 | Cisco Technology, Inc. | Methods and apparatus for providing and obtaining resource usage information |
US6445696B1 (en) | 2000-02-25 | 2002-09-03 | Network Equipment Technologies, Inc. | Efficient variable rate coding of voice over asynchronous transfer mode |
WO2001067672A2 (en) * | 2000-03-07 | 2001-09-13 | Sun Microsystems, Inc. | Virtual channel flow control |
US7142558B1 (en) | 2000-04-17 | 2006-11-28 | Cisco Technology, Inc. | Dynamic queuing control for variable throughput communication channels |
US6466541B1 (en) | 2000-05-31 | 2002-10-15 | Fujitsu Network Communications, Inc. | Cell pacing on a network link employing a rate-based flow control protocol with underlying credit-based flow control mechanisms |
US6452903B1 (en) | 2000-05-31 | 2002-09-17 | Fujitsu Network Communications, Inc. | Network switch supporting rate-based and credit-based flow control mechanisms on a link-by-link basis |
US6987732B2 (en) * | 2000-12-15 | 2006-01-17 | Tellabs San Jose, Inc. | Apparatus and methods for scheduling packets in a broadband data stream |
US7116639B1 (en) | 2000-12-21 | 2006-10-03 | International Business Machines Corporation | System and method for determining network discrete utilization |
US6996064B2 (en) * | 2000-12-21 | 2006-02-07 | International Business Machines Corporation | System and method for determining network throughput speed and streaming utilization |
US7193966B2 (en) * | 2001-06-25 | 2007-03-20 | Telefonakitebolaget Lm Ericsson (Publ) | Triggered packet data rate change in a communication system |
US7042841B2 (en) * | 2001-07-16 | 2006-05-09 | International Business Machines Corporation | Controlling network congestion using a biased packet discard policy for congestion control and encoded session packets: methods, systems, and program products |
US7218610B2 (en) | 2001-09-27 | 2007-05-15 | Eg Technology, Inc. | Communication system and techniques for transmission from source to destination |
US20050152397A1 (en) * | 2001-09-27 | 2005-07-14 | Junfeng Bai | Communication system and techniques for transmission from source to destination |
US7349406B1 (en) * | 2001-10-18 | 2008-03-25 | Network Equipment Technologies Inc. | Method and apparatus for virtual network connection merging |
US7103350B2 (en) * | 2001-11-16 | 2006-09-05 | Nortel Networks Limited | Scheduler with fairness control and quality of service support |
CA2482130A1 (en) * | 2002-04-12 | 2003-10-23 | Siemens Aktiengesellschaft | Method for common control of the bandwidths of a group of individual information flows |
US7154853B2 (en) * | 2002-05-02 | 2006-12-26 | Intel Corporation | Rate policing algorithm for packet flows |
US7296083B2 (en) * | 2002-06-28 | 2007-11-13 | Microsoft Corporation | Method and system for measuring load and capacity on a variable capacity channel |
US6919004B2 (en) * | 2002-07-02 | 2005-07-19 | E. I. Du Pont De Nemours And Company | Cathodic electrocoating composition containing a morpholine dione crosslinking agent |
US7408876B1 (en) | 2002-07-02 | 2008-08-05 | Extreme Networks | Method and apparatus for providing quality of service across a switched backplane between egress queue managers |
US7286552B1 (en) * | 2002-07-26 | 2007-10-23 | Extreme Networks | Method and apparatus for providing quality of service across a switched backplane for multicast packets |
US7599292B1 (en) | 2002-08-05 | 2009-10-06 | Extreme Networks | Method and apparatus for providing quality of service across a switched backplane between egress and ingress queue managers |
US20070109969A1 (en) * | 2003-04-07 | 2007-05-17 | Markus Baumeister | Method of ensuring the quality of service in a network |
US7801038B2 (en) * | 2003-07-14 | 2010-09-21 | Siemens Corporation | Method and apparatus for providing a delay guarantee for a wireless network |
GB2405740B (en) * | 2003-09-03 | 2005-08-31 | Hanpin Electron Co Ltd | Dual-use audio signal player for analogue record and digital CD |
US7085228B2 (en) * | 2003-10-31 | 2006-08-01 | Interdigital Technology Corp. | Adaptive radio resource management for wireless local area networks |
US7466996B2 (en) * | 2003-12-22 | 2008-12-16 | International Business Machines Corporation | Global management of local link power consumption |
US20050163048A1 (en) * | 2004-01-07 | 2005-07-28 | Amit Arora | Method and system for providing committed information rate (CIR) based fair access policy |
US7415470B2 (en) * | 2004-08-12 | 2008-08-19 | Oracle International Corporation | Capturing and re-creating the state of a queue when migrating a session |
US7502824B2 (en) * | 2004-08-12 | 2009-03-10 | Oracle International Corporation | Database shutdown with session migration |
US7434262B2 (en) * | 2004-12-08 | 2008-10-07 | At&T Intellectual Property I, L.P. | Methods and systems that selectively resurrect blocked communications between devices |
DE102004062116B3 (en) * | 2004-12-23 | 2006-05-11 | Ab Skf | Bearing arrangement for computer tomography has bearing with inner ring, which stores construction unit, and outer ring, which is connected with damping element, fitted as single element and contain hollow cylindrical basic outline |
US9176772B2 (en) * | 2005-02-11 | 2015-11-03 | Oracle International Corporation | Suspending and resuming of sessions |
US8547843B2 (en) * | 2006-01-20 | 2013-10-01 | Saisei Networks Pte Ltd | System, method, and computer program product for controlling output port utilization |
US20070237074A1 (en) * | 2006-04-06 | 2007-10-11 | Curry David S | Configuration of congestion thresholds for a network traffic management system |
US7623449B2 (en) * | 2006-05-14 | 2009-11-24 | Atrica Israel Ltd. | Policy aware frame loss measurement |
US20080056146A1 (en) * | 2006-08-29 | 2008-03-06 | Elliott Steven L | Method and apparatus for determining maximum round trip times for a network socket |
US8014308B2 (en) * | 2006-09-28 | 2011-09-06 | Microsoft Corporation | Hardware architecture for cloud services |
FR2909503B1 (en) * | 2006-12-04 | 2009-10-09 | Alcatel Sa | METHOD OF ESTABLISHING A BIDIRECTIONAL CONNECTION |
US20090010180A1 (en) * | 2007-07-03 | 2009-01-08 | Qualcomm Incorporated | Methods and apparatus for resource provisioning and planning in a communication network |
US20090144404A1 (en) * | 2007-12-04 | 2009-06-04 | Microsoft Corporation | Load management in a distributed system |
US20090150536A1 (en) * | 2007-12-05 | 2009-06-11 | Microsoft Corporation | Application layer congestion control |
US8050177B2 (en) * | 2008-03-31 | 2011-11-01 | Intel Corporation | Interconnect bandwidth throttler |
US8531961B2 (en) | 2009-06-12 | 2013-09-10 | Cygnus Broadband, Inc. | Systems and methods for prioritization of data for intelligent discard in a communication network |
WO2010144833A2 (en) | 2009-06-12 | 2010-12-16 | Cygnus Broadband | Systems and methods for intelligent discard in a communication network |
US8627396B2 (en) * | 2009-06-12 | 2014-01-07 | Cygnus Broadband, Inc. | Systems and methods for prioritization of data for intelligent discard in a communication network |
US8549038B2 (en) * | 2009-06-15 | 2013-10-01 | Oracle International Corporation | Pluggable session context |
US9270580B1 (en) | 2009-08-28 | 2016-02-23 | Brocade Communications Systems, Inc. | Method and system for traffic isolation in a network |
US8885657B2 (en) * | 2009-11-06 | 2014-11-11 | Brocade Communications Systems, Inc. | Automatic switch port selection |
US9391849B2 (en) * | 2009-11-06 | 2016-07-12 | Brocade Communications Systems, Inc. | Back pressure remediation |
US8593964B1 (en) | 2009-11-06 | 2013-11-26 | Brocade Communications Systems, Inc. | Method and system for traffic management |
US8891368B2 (en) * | 2009-11-06 | 2014-11-18 | Brocade Communications Systems, Inc. | Presentation of a selected port |
US8910839B2 (en) * | 2010-01-18 | 2014-12-16 | Safariland, Llc | Locking device safety mechanism and related holster assembly |
KR101212645B1 (en) * | 2011-04-26 | 2012-12-14 | 서울대학교산학협력단 | Method and Apparatus for resource allocation in virtual network environment |
US9276953B2 (en) | 2011-05-13 | 2016-03-01 | International Business Machines Corporation | Method and apparatus to detect and block unauthorized MAC address by virtual machine aware network switches |
US8670450B2 (en) | 2011-05-13 | 2014-03-11 | International Business Machines Corporation | Efficient software-based private VLAN solution for distributed virtual switches |
US8837499B2 (en) | 2011-05-14 | 2014-09-16 | International Business Machines Corporation | Distributed fabric protocol (DFP) switching network architecture |
US20120291034A1 (en) | 2011-05-14 | 2012-11-15 | International Business Machines Corporation | Techniques for executing threads in a computing environment |
US20120287785A1 (en) | 2011-05-14 | 2012-11-15 | International Business Machines Corporation | Data traffic handling in a distributed fabric protocol (dfp) switching network architecture |
US9497073B2 (en) | 2011-06-17 | 2016-11-15 | International Business Machines Corporation | Distributed link aggregation group (LAG) for a layer 2 fabric |
US8767529B2 (en) | 2011-09-12 | 2014-07-01 | International Business Machines Corporation | High availability distributed fabric protocol (DFP) switching network architecture |
US20130064066A1 (en) | 2011-09-12 | 2013-03-14 | International Business Machines Corporation | Updating a switch software image in a distributed fabric protocol (dfp) switching network |
US8750129B2 (en) | 2011-10-06 | 2014-06-10 | International Business Machines Corporation | Credit-based network congestion management |
US9065745B2 (en) | 2011-10-06 | 2015-06-23 | International Business Machines Corporation | Network traffic distribution |
US9894397B2 (en) | 2011-12-15 | 2018-02-13 | Cisco Technology, Inc. | Controlling bit-rates for media streaming sessions |
US9455794B2 (en) | 2012-10-29 | 2016-09-27 | Qualcomm Incorporated | Device registration and sounding in a time-division multiple access network |
US9699088B2 (en) * | 2014-11-10 | 2017-07-04 | Hughes Network Systems, Llc | Service plan based flow control |
US10289617B2 (en) | 2015-12-17 | 2019-05-14 | Oracle International Corporation | Accessing on-premise and off-premise datastores that are organized using different application schemas |
US10387387B2 (en) | 2015-12-17 | 2019-08-20 | Oracle International Corporation | Enabling multi-tenant access to respective isolated data sets organized using different application schemas |
US10303894B2 (en) | 2016-08-31 | 2019-05-28 | Oracle International Corporation | Fine-grained access control for data manipulation language (DML) operations on relational data |
CN112235209A (en) * | 2019-07-25 | 2021-01-15 | 北京天德科技有限公司 | Intra-network control of virtual circuits |
CN112437019B (en) * | 2020-11-30 | 2022-01-25 | 中国人民解放军国防科技大学 | Active transmission method based on credit packet for data center |
CN114124841A (en) * | 2021-11-19 | 2022-03-01 | 青岛信大云谷信息科技有限公司 | Method for controlling high-speed forwarding data lightweight flow |
Family Cites Families (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5377327A (en) * | 1988-04-22 | 1994-12-27 | Digital Equipment Corporation | Congestion avoidance scheme for computer networks |
US5018137A (en) * | 1988-06-27 | 1991-05-21 | Digital Equipment Corporation | Transparent load sharing for parallel networks |
JP2753294B2 (en) * | 1988-12-23 | 1998-05-18 | 株式会社日立製作所 | Packet congestion control method and packet switching device |
GB8914983D0 (en) * | 1989-06-29 | 1989-08-23 | Digital Equipment Int | Congestion control in computer networks |
US5193151A (en) * | 1989-08-30 | 1993-03-09 | Digital Equipment Corporation | Delay-based congestion avoidance in computer networks |
US5029164A (en) * | 1990-04-13 | 1991-07-02 | Digital Equipment Corporation | Congestion avoidance in high-speed network carrying bursty traffic |
EP0487235B1 (en) * | 1990-11-21 | 1999-02-03 | AT&T Corp. | Bandwidth and congestion management in accessing broadband ISDN networks |
US5339313A (en) * | 1991-06-28 | 1994-08-16 | Digital Equipment Corporation | Method and apparatus for traffic congestion control in a communication network bridge device |
US5274644A (en) * | 1991-11-05 | 1993-12-28 | At&T Bell Laboratories | Efficient, rate-base multiclass access control |
US5390299A (en) * | 1991-12-27 | 1995-02-14 | Digital Equipment Corporation | System for using three different methods to report buffer memory occupancy information regarding fullness-related and/or packet discard-related information |
US5313454A (en) * | 1992-04-01 | 1994-05-17 | Stratacom, Inc. | Congestion control for cell networks |
US5381404A (en) * | 1992-07-14 | 1995-01-10 | Mita Industrial Co., Ltd. | Packet-switching communication network and method of design |
JPH0646082A (en) * | 1992-07-22 | 1994-02-18 | Toshiba Corp | Information transfer control system |
US5390173A (en) * | 1992-10-22 | 1995-02-14 | Digital Equipment Corporation | Packet format in hub for packet data communications system |
US5359593A (en) * | 1993-08-26 | 1994-10-25 | International Business Machines Corporation | Dynamic bandwidth estimation and adaptation for packet communications networks |
US5457687A (en) * | 1993-09-02 | 1995-10-10 | Network Equipment Technologies, Inc. | Method and apparatus for backward explicit congestion notification (BECN) in an ATM network |
US5426635A (en) * | 1993-09-08 | 1995-06-20 | At&T Corp. | Method for adaptive control of windows and rates in networks |
-
1995
- 1995-11-15 ZA ZA959722A patent/ZA959722B/en unknown
- 1995-11-28 JP JP30960095A patent/JP3833739B2/en not_active Expired - Fee Related
- 1995-12-01 AT AT95118918T patent/ATE218259T1/en not_active IP Right Cessation
- 1995-12-01 ES ES95118918T patent/ES2177600T3/en not_active Expired - Lifetime
- 1995-12-01 EP EP95118918A patent/EP0719012B1/en not_active Expired - Lifetime
- 1995-12-01 DE DE69526839T patent/DE69526839T2/en not_active Expired - Lifetime
- 1995-12-05 CA CA002164489A patent/CA2164489A1/en not_active Abandoned
- 1995-12-05 AU AU40202/95A patent/AU703410B2/en not_active Ceased
- 1995-12-06 CN CN95120526A patent/CN1111999C/en not_active Expired - Fee Related
- 1995-12-13 AR AR33458595A patent/AR000331A1/en unknown
- 1995-12-14 BR BR9505887A patent/BR9505887A/en not_active Application Discontinuation
- 1995-12-15 KR KR1019950050381A patent/KR960027839A/en not_active Application Discontinuation
-
1996
- 1996-01-11 US US08/585,265 patent/US5633861A/en not_active Expired - Lifetime
Also Published As
Publication number | Publication date |
---|---|
CN1111999C (en) | 2003-06-18 |
EP0719012A2 (en) | 1996-06-26 |
US5633861A (en) | 1997-05-27 |
EP0719012A3 (en) | 1997-06-11 |
ATE218259T1 (en) | 2002-06-15 |
ZA959722B (en) | 1996-05-31 |
CN1137717A (en) | 1996-12-11 |
JP3833739B2 (en) | 2006-10-18 |
DE69526839T2 (en) | 2002-09-12 |
KR960027839A (en) | 1996-07-22 |
AU4020295A (en) | 1996-06-27 |
EP0719012B1 (en) | 2002-05-29 |
AR000331A1 (en) | 1997-06-18 |
ES2177600T3 (en) | 2002-12-16 |
JPH08237301A (en) | 1996-09-13 |
AU703410B2 (en) | 1999-03-25 |
DE69526839D1 (en) | 2002-07-04 |
BR9505887A (en) | 1998-01-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
AU703410B2 (en) | Traffic management and congestion control for ATM | |
US7054269B1 (en) | Congestion control and traffic management system for packet-based networks | |
US5757770A (en) | Method and device for the control of congestion in sporadic exchanges of data packets in a digital transmission network | |
US5267232A (en) | Method of controlling data transmission in ATM network with cell loss priority level | |
US5313454A (en) | Congestion control for cell networks | |
EP0577359B1 (en) | Congestion control in high speed networks | |
EP0784895B1 (en) | Flow control method and apparatus for cell-based communication networks | |
US5835484A (en) | Method and apparatus for controlling congestion in communication network | |
EP1798915B1 (en) | Packet forwarding device avoiding packet loss of out of profile packets in the shaper by remarking and redirecting the packet to a lower priority queue | |
US4769810A (en) | Packet switching system arranged for congestion control through bandwidth management | |
US7046631B1 (en) | Method and apparatus for provisioning traffic dedicated cores in a connection oriented network | |
US6934296B2 (en) | Packet transfer device and packet transfer method adaptive to a large number of input ports | |
EP0800294B1 (en) | Method to control data flow rate, queuing network node and packet switching network | |
CA2272221C (en) | Method for monitoring and controlling traffic in real time in an atm switching node | |
Yin et al. | Analysis of the leaky bucket algorithm for on-off data sources | |
US5956322A (en) | Phantom flow control method and apparatus | |
US6597662B1 (en) | Apparatus and method for optimizing max-min fair rate control in ABR sessions | |
US5978357A (en) | Phantom flow control method and apparatus with improved stability | |
US7218608B1 (en) | Random early detection algorithm using an indicator bit to detect congestion in a computer network | |
CA2285086C (en) | Improved phantom flow control method and apparatus | |
KR100319457B1 (en) | Traffic control method for ATM switches | |
WO1998043395A9 (en) | Improved phantom flow control method and apparatus | |
US7142558B1 (en) | Dynamic queuing control for variable throughput communication channels | |
JP3203610B2 (en) | Cell flow control device | |
Lin et al. | End-to-end data transport using SSCOP in ATM networks |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
EEER | Examination request | ||
FZDE | Discontinued |