WO1997037310A1 - Performance monitoring of an atm switch - Google Patents

Performance monitoring of an atm switch Download PDF

Info

Publication number
WO1997037310A1
WO1997037310A1 PCT/US1997/004999 US9704999W WO9737310A1 WO 1997037310 A1 WO1997037310 A1 WO 1997037310A1 US 9704999 W US9704999 W US 9704999W WO 9737310 A1 WO9737310 A1 WO 9737310A1
Authority
WO
WIPO (PCT)
Prior art keywords
packet
management
network connection
delay
data packets
Prior art date
Application number
PCT/US1997/004999
Other languages
French (fr)
Inventor
Thomas M. Chen
Stephen S. Liu
Original Assignee
Gte Laboratories Incorporated
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Gte Laboratories Incorporated filed Critical Gte Laboratories Incorporated
Priority to JP53541997A priority Critical patent/JP3894232B2/en
Priority to CA002250278A priority patent/CA2250278C/en
Priority to EP97917052A priority patent/EP0892958A4/en
Publication of WO1997037310A1 publication Critical patent/WO1997037310A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0852Delays
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/10Active monitoring, e.g. heartbeat, ping or trace-route
    • H04L43/106Active monitoring, e.g. heartbeat, ping or trace-route using time related information in packets, e.g. by adding timestamps
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/50Testing arrangements
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/20Support for services
    • H04L49/205Quality of Service based
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/30Peripheral units, e.g. input or output ports
    • H04L49/3009Header conversion, routing tables or routing tags
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/54Store-and-forward switching systems 
    • H04L12/56Packet switching systems
    • H04L12/5601Transfer mode dependent, e.g. ATM
    • H04L2012/5619Network Node Interface, e.g. tandem connections, transit switching
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/54Store-and-forward switching systems 
    • H04L12/56Packet switching systems
    • H04L12/5601Transfer mode dependent, e.g. ATM
    • H04L2012/5625Operations, administration and maintenance [OAM]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00Data switching networks
    • H04L12/54Store-and-forward switching systems 
    • H04L12/56Packet switching systems
    • H04L12/5601Transfer mode dependent, e.g. ATM
    • H04L2012/5638Services, e.g. multimedia, GOS, QOS
    • H04L2012/5646Cell characteristics, e.g. loss, delay, jitter, sequence integrity
    • H04L2012/5652Cell construction, e.g. including header, packetisation, depacketisation, assembly, reassembly

Definitions

  • the present invention relates to electronic communications networks and. in particular, to performance monitoring of packet-switched data nerworks.
  • the Broadband Integrated Services Digital Network can accommodate a variety of packet-o ⁇ ented protocols, including asynchronous transfer mode (ATM), in which standardized 53-byte data packets are utilized [ITU-T Ree. 1.121. Broadband Aspects of ISDN. Melbourne, Nov. 1988].
  • ATM and some other packet-o ⁇ ented protocols are connection-oriented, utilizing what are known in ATM as virtual path connections (VPC) and virtual channel connections (VCC) that stretch between two endpoints across transmission links connected by intermediate nodes. These intermediate nodes are devices that forward and direct the packets to their intended destination endpoint. i.e. routers, or switches.
  • Packet-switched data networks are capable of supporting a simultaneous mix of voice, video, and data services over either constant bit rate (CBR) or variable bit rate (VBR) connections.
  • CBR constant bit rate
  • VBR variable bit rate
  • an ATM network can guarantee to the customer a particular level of performance, known as a quality of service (QoS). that is negotiated between the user and the network during connection set-up. Once agreed upon, the QoS can only be changed through renegotiation between the two parties.
  • QoS in ATM is specified mainly in terms of acceptable ceil delay and acceptable cell loss rate [ATM Forum, ATM User-Network Interface Specification Version 3.
  • ATM supports multiple QoS classes geared to differing application requirements for delay and loss performance.
  • a VCC or VPC is therefore characte ⁇ zed by its requested QoS and throughput parameters.
  • a VPC may carry statistically multiplexed VCCs of diffe ⁇ ng QoS classes, but the overall QoS of the VPC must satisfy the most demanding QoS of the earned VCCs Specified QoS parameter values may be either explicitly requested by the user or implicitly associated with a particular service.
  • Performance management as a function of network management, has the simultaneous responsibilities of maintaining both the QoS per virtual connection and the efficient utilization of the network. Performance management itself depends largely on the data obtained through performance monitoring, which is conducted through measurement of the QoS per VPC/VCC. Because of the use of statistical multiplexing in ATM networks, the QoS of a virtual connection may in particular be adversely affected by the random flow of traffic on the network.
  • QoS Quality of Service
  • Network providers need to be able to detect and diagnose troubles quickly in order to minimize the effect of those troubles on network users.
  • Performance monitoring can be done either through out-of-service testing or in-service QoS monitoring.
  • Out-of-service testing involves performance measurements made under controlled traffic conditions. This type of testing has a number of drawbacks, not the least of which is that service on the network must be :nterrupted. In addition. Lie results received are only as accurate as the model used to create the controlled traffic conditions. It is entirely possible that the controlled conditions tested may not correctly model some of the conditions actually seen by tne network du ⁇ ng service. In addition, this method does not allow rapid detection of any subsequent degradation of the network.
  • In-service QoS monitoring involves continuous performance measurements on actual user traffic. No service interruption occurs and network degradation is instantaneously detected. Continuous performance monitoring is pa ⁇ icuiarly important in high-speed networks where service deterioration may quickly affect a large amount of user data. In-service QoS measurements are also useful to supplement traffic control algorithms based on assumed traffic models, such as those used in out-of-service testing, that may not accurately characterize the actual traffic.
  • OAM Operations and Maintenance
  • ITU-T Ree. 1.610 B-ISDN Operations and Maintenance Principles and Functions, Geneva, June 1992; Bellcore. Generic Requirements for Operations of Broadband Switching Systems, TA-NWT00 1248. issue 2. Oct. 1993 ].
  • OAM cells are specialized versions of standard 53 -byte ATM cells, having a normal 5-byte header with a 48-byte L ⁇ formation field, but carrying management information within the function- specific information fields instead of user data.
  • ATM user cells themselves cannot be used for QoS monitoring because the payload is full of user data and no header fields are available to receive the information that is needed by the performance monitoring system for calculation and storage of cell loss and cell delay.
  • a 4- bit OAM Function Type field in the information field further indicates the specific performance management function performed by the OAM cell, such as forward monitoring (0000), backward reporting (0001), or both (0010).
  • a 10-bit EDC field contains an error detection code that is computed over the information field of the cell.
  • the forward monitoring procedure consists of inserting OAM performance management cells into the cell stream between blocks of user cells at one endpoint of the virtual connection.
  • Backward reporting involves putting observed performance data into an OAM cell at a virtual connection endpoint and then sending the OAM cell backwards in the cell stream direction from which it came, in order to report the performance of that particular virtual connection to the originating endpoint.
  • the OAM cell contains the number of user cells sent in the preceding block (Total User Cell Number) in a 2-byte function-specific field.
  • the destination endpoint of the virtual path counts the number of user cells received between the arrival of the previous OAM cell and the current OAM cell. If fewer cells are received at the destination than are indicated in the current OAM cell, the difference is interpreted to be the number of lost cells.
  • the major disadvantage of the current network performance monitoring method for measuring cell loss is that it cannot be assured of actually determining the exact number of cells lost. This is because cells can also be misinserted into the cell stream. Therefore, for example, if the OAM cell indicates that N user cells were sent and then N- 1 cells are actually received, it is not possible to be sure whether 1 cell was lost, or whether 2 cells were lost and 1 misinserted, and so on. It is also not possible, using this method, to determine from where along the virtual connection, or from where within a cell block, the cell loss actually occurred.
  • the OAM performance management cell contains an optional 4-byte Timestamp field that is used to indicate the time that the OAM cell was inserted into the user stream.
  • the timestamp is then compared with the arrival time of the cell at the virtual connection endpoint and the difference is inferred to be the end-to-end delay experienced by a cell along the virtual connection.
  • this type of timestamp is meaningful at the destination only if the clocks at the two endpoints of the virtual connection are synchronized to the same time-of-day. something which is generally not practical.
  • the existing method of delay measurement does not take into account the fact that end-to-end delay actually consists of two distinct types of delay: fixed propagation delay, which is that delay introduced into the connection at a predictable level by the transmission links extending between the network nodes, and random delay.
  • Random delay is that delay which is introduced by each of the intermediate nodes themselves, and can vary widely with network traffic levels and switch buffer capacity. In general, this type of delay is primarily due to queuing delays, which tend to increase with increased traffic load.
  • the existing method of performance monitoring using OAM cells is not able to track the delay that is actually experienced at each node along the virtual connection due to such factors.
  • a timestamp allows computation of the round- trip delay back at the originating endpoint, without the need for time-of-day clock synchronization between nodes.
  • the one-way end-to-end delay is then estimated as half the round-trip delay.
  • the loopback cell method can measure the round-trip delays without need for clock synchronization, it gives only an estimate of the one-way cell delay. This estimate may in actuality be very inaccurate if the network congestion or characteristics have altered greatly between the two halves of the OAM cell's trip.
  • this method is also not able to measure the delay occurring at individual intermediate nodes.
  • a new class of packet which has a payload that is modified by each node along a virtual connection.
  • Management packets are used to accurately measure and report the end-to-end QoS along a virtual connection, as well as to measure and report specific performance parameters at individual intermediate nodes
  • management cells are implemented by defining an entirely new type of cell or by defining additional fields within existing OAM cells.
  • the switch at each intermediate node modifies the payload of management packets and locally measures delay and packet loss.
  • a function type field specifies the particular performance monitoring function performed by each management packet.
  • the information field of the management packet optionally contains timestamp fields, delay-stamp fields, and packet loss count fields.
  • intermediate switches can measure the difference between the arrival time and the departure time of any packet at that switch.
  • a delay-stamp field is incremented by the local delay measured at each switch.
  • the delay-stamp field indicates the cumulative amount of time that the packet has been delayed by the nodes of the network.
  • a node-by-node delay measurement management packet allows computation and relay of instantaneous delay information by utilizing multiple delay-stamp fields to record each delay encountered on a node by-node basis.
  • the end-to-end delay is computed as the sum of the recorded delay-stamp fields.
  • the switch may record in one timestamp field the time that the packet arrives, and then record the time the packet leaves into another timestamp field, with the calculation of node-by-node delay made when the packet reaches the endpoint of the virtual connection.
  • intermediate switches can count the number of packets each discards.
  • a node receives an end-to-end loss accumulating management packet, it increments a packet loss count field by the number of packets discarded by the node within that virtual connection since the previous loss management packet arrived.
  • the packet loss count field reflects the total number of discarded packets in the preceding block of user packets.
  • Node-by-node loss measurement management packets contain multiple packet loss count fields for recording the number of discarded packets counted at each node since the departure of the previous loss management packet in that virtual connection. The total packet loss is then calculated at the endpoint of the virtual connection.
  • Associated management packets experience the same delay and packet loss rate as the user data packets in a particular virtual connection, implemented in one embodiment of the invention by having the management packets follow the same path as the user packets through the switch-
  • the pay loads of associated management packets are processed without the packet being removed from the data packet stream.
  • this is implemented by using the switch internal routing header.
  • the switch input header processor uses the network routing instructions for each packet to create an internal routing header that, for a delay management packet, includes the time of entry into the switch.
  • the time a delay management packet arrives at the switch is recorded directly into a timestamp field in the packet while the previous user data packet is being sent to the switch fabric.
  • all the data packets are routed through the switch fabric and directed towards their intended output ports through an output header processor, which removes the internal routing header from the packets, compares the time-of-day stamp in the intemal routing header of delay management packets to the current time-of-day to calculate the delay incurred by each packet in the switch, and records the difference in the delay-stamp field of the management data packet. This occurs while the previous user data packet in the datastream is being sent to the output port so that the management data packet retains its place in the datastream. A similar procedure is followed if the timestamp field is used, with the output header processor reading the time of entry from that field.
  • packet loss counts are recorded in loss management packets.
  • the processor records into the packet loss count field the number of packets discarded by the switch within that particular virtual connection since the last loss management packet was processed.
  • the packet loss count is recorded while the previous user data packet in the datastream is being sent to the output port, allowing the loss management data packet to retain its place in the datastream.
  • Non-associated management packets are extracted from the datastream and diverted to a special processor, therefore not retaining their positions relative to the user data packets in the same virtual connection.
  • Non-associated management packets are used in one aspect of the invention by a control processor to report cumulative performance or usage measurements.
  • the input header processor creates an internal routing header for non-associated management packets that directs the switch fabric to route the packets to the control processor for modification.
  • Non-associated management packets are used in another aspect of the invention to implement local measurement of packet delay and packet losses by replacing the non-associated management packet temporarily with a marker packet that holds the place of the non-associated management packet in the datastream while the control processor records the delay or packet loss count information in the payload of the management packet. The marker packet is then replaced in the datastream with the modified non-associated management packet.
  • Local measurement of delay and packet loss is implemented in an alternate embodiment of the invention by routing all data packets through a control processor, recording delay and packet loss count information into appropriate management packets, and performing dummy writes to user data packets.
  • Fig. l is a block diagram of an embodiment of the management packet of the present invention for an ATM implementation; Fig.
  • FIG. 2 is a block diagram illustrating one embodiment of the method of the present invention, utilizing end-to-end delay accumulating management cells
  • Fig. 3 is a block diagram illustrating another embodiment of the method of the present invention, utilizing node-by-node delay measurement management cells:
  • Fig. 4 is a block diagram illustrating an embodiment of the method of the present invention, utilizing end-to-end loss accumulating management cells:
  • Fig. 5 is a block diagram illustrating another embodiment of the method of the present invention utilizing node-by-node loss measurement management cells;
  • Fig. 6 is a block diagram illustrating the flow of associated management packets through a switch with local observability of packet delay according to one aspect of the present invention; and
  • Fig. 7 is a block diagram illustrating the flow of non-associated management packets through a switch according to another aspect of the present invention.
  • management packet having a payload that can be modified by each node along a virtual connection.
  • This method can be applied to any packet-switching network.
  • these "management cells” can optionally be implemented as extensions of current OAM cells.
  • management cells carry measurement data along the virtual connection in the cell information field.
  • the major difference between management cells and prior art OAM cells is that the information field of a management cell is modified by all the network nodes along a virtual connection, not just by the virtual connection endpoints. This new capability makes it possible for the performance monitoring system to track the performance of each node along the virtual connection, rather than just the end- to-end performance of the virtual connection as a whole.
  • Management packets are used to accurately measure the end-to-end QoS experienced by the user packets along any virtual connection.
  • specific performance parameters measured at individual intermediate nodes are made available to the performance monitoring system. This allows in-service monitoring to be used to accurately identify the QoS actually seen by users and to detect exact causes of performance deterioration before users are seriously affected.
  • the method of the present invention therefore offers vastly improved in-service monitoring of both end-to-end packet delay and packet loss rate.
  • the method requires that the router or switch at each node have specific, unique capabilities, including local processing measurement of packets and local measurement of packet delay and packet loss.
  • Management packets are used to accurately measure end-to-end packet delay, without the need for time-of-day clock synchronization between network connection endpoints, by measuring and reporting the packet delay at each node along a virtual connection. Likewise. management packets allow accurate measurement and reporting of both end-to-end packet loss and packet loss at each node along a virtual connection.
  • the ATM management cells of the preferred embodiment may be implemented either by defining entirely new cells, or by defining additional fields within existing OAM cells, since existing OAM cells used for performance management have 33 bytes in the information field that are unused.
  • An embodiment of a management cell for an ATM application is shown in Fig. 1.
  • entirely new VCI and PT codes are defined that uniquely identify the cell to the network as an ATM management cell.
  • the management cell is instead uniquely identified as a management cell to the network by a special code in the 4-bit OAM Cell Type field 4.
  • a 4-bit Function Type field 6 further indicates the specific performance monitoring function performed by the management cell.
  • a 10-bit EDC field 8 contains an error detection code that is computed over the information field of the cell.
  • the Function Type field 6 contains a code that describes to the network the type of performance monitoring function performed by the ATM management cell, labeling the cell as an "end-to-end delay accumulating" cell, a “node-by-node delay measurement” cell, an "end-to-end loss accumulation” cell, a “node-by-node loss measurement” cell, or as some combination of the previous types. Additional codes can be defined for the monitoring of other performance parameters.
  • the information field of the management cell will also optionally contain a 4-byte Timestamp field 9, a Delay- stamp field 10, and a Cell Loss Count field 12, as well as other function-specific fields 14 that are used for additional delay-stamps, cell loss counts, etc., as needed. These fields can also be used for monitoring and reporting other performance parameters.
  • a switch For measurement of packet delay, a switch is required that can measure the difference between the arrival time and the departure time of any packet at that switch, a capability known as "local observability" of packet delay. Because most switches have at least some buffering capability, the delay incurred by a packet in the switch can be highly variable, particularly where a priority-based buffering scheme is implemented. Packet delay within a switch tends to increase as network traffic increases. Existing packet switches are not currently implemented so as to have the ability to measure the delay experienced by a packet within the switch itself
  • the management packet method utilizes measurements of the individual packet delay experienced between the input and output of each of the intermediate switches.
  • Existing switches used with performance monitoring OAM cells have the ability to timestamp OAM cells when they enter a cell stream. In the management packet method, this activity can be replicated at each intermediate node. If the management packet is an end-to-end delay accumulating management cell, the intermediate node takes the Delay-stamp field 10. which was set to zero at the originating endpoint of the virtual connection, and increments it by the local delay measured at that switch, as illustrated in Fig. 2 for an ATM embodiment.
  • the Delay-stamp field 10 initially contains a delay TO from previous intermediate nodes.
  • the time the cell of Fig. 1 enters the node 16 is recorded in the Timestamp field 9. This field is then compared to the time the cell leaves the node and the Delay-stamp field 10 is incremented by the difference. Tl .
  • a new time in is recorded in the Timestamp field 9, corresponding to the time the cell has entered the node 18.
  • This field is again compared to the time the cell exits the node 18 and the difference, T2 is used to increment the value in the Delaystamp field 10.
  • T2 is used to increment the value in the Delaystamp field 10.
  • a similar procedure is followed at node 3 20. and so on, until the endpoint of the virtual connection is reached.
  • the time a packet enters each node can be incorporated into a field in -13-
  • the switch then examines the time of entry recorded in the routing header to calculate the delay that is written or incremented into the Delay-stamp field.
  • the Delay-stamp field 10 indicates the cumulative amount of time that the cell has been variably delayed by the nodes of the network. If desired, this delay can then be added to the value of fixed propagation delay caused by the transmission links between the nodes in order to arrive at the total time that the cell has spent in the network.
  • a node-by-node delay measurement management packet may be used.
  • a particular advantage of the node- by-node delay measurement management packet is that it allows the computation and relay of instantaneous delay information.
  • This type of management packet contains multiple delay-stamp fields for recording each local delay encountered on a node-by-node basis, as illustrated in Fig. 3 for an ATM embodiment. In Fig. 3.
  • the cell is stamped in the Timestamp field O, or in the separate internal routing header, with the time of entry into the node. This time is then compared to the time of exit from that node and the calculated delay is recorded into the cell in one of a multiple of unused Delay-stamp fields 22 This allows monitoring of the delay performance of each intermediate node at the virtual connection endpoint. If desired, the end-to-end cell delay can also be computed at the destination endpoint, being the sum of the delay-stamp fields. Alternatively, additional timestamp fields can be used for each node, allowing the switch to merely record in one timestamp field the time that the packet arrives, and then record the time the packet leaves into another timestamp field.
  • node-by-node and end-to-end delay can then be made when the packet has reached the endpoint of the virtual connection.
  • the cell is for the cell to be a combination delay cell, allowing the switch to record both the accumulated and the node-by-node delays in the same packet. Measurement of packet loss requires a switch with local observability of packet loss, allowing the switch to count the number of packets it discards per virtual connection.
  • Packet discards occur for various reasons, such as switch buffer overflow due to excessive traffic, corrupted routing information in a cell header field or on the connection, and when packets have failed to arrive at their destination within a pre-specified time and have therefore "expired.”
  • Each switch must count the number of packets it discards within each virtual connection. This capability is already available in certain existing switches, such as the AT&T GlobeView-2000 switch [See AT&T, Globe View-2000 Broadband System Network: Network Management, 255-179-079, Issue 1.0, Release 1.3, 1995] and the Newbridge 36150 MainStreet switch [See Newbridge, 36150 MainStreet
  • the management packet used to measure packet loss can be the same packet as used to measure packet delay or a different packet.
  • Packet loss is measured according to one embodiment of the present invention through the insertion of end-to-end loss accumulating management packets into a virtual connection between blocks of user packets, as illustrated in Fig. 4 for an ATM embodiment.
  • Each end-to-end loss accumulating management cell contains a Cell Loss Count field 12 that is initialized to zero.
  • node 24 receives the end-to- end loss accumulating management cell with a Cell Loss Count field value NO, it increments the Cell Loss Count field 12 by the number of cells discarded by the node 24 in that virtual connection since the previous loss management cell arrived in that particular virtual connection.
  • the cell is then received by node 226, which increments the Cell Loss Count field 12 by the number of cells discarded within that virtual connection by that node.
  • a similar procedure is followed at the next node. node 3 28. and so on.
  • the Cell Loss Count field 12 accurately reflects the total number of discarded cells in the preceding block of user cells. In contrast to the use of existing OAM performance management cells, it is not required that a node know how many cells to expect in the user cell block, since each node accurately counts how many cells it discards.
  • node-by-node loss measurement management packets containing multiple fields to record the number of discarded packets counted at each node may be used. An illustration of this is shown in Fig. 5 for an ATM embodiment.
  • the number of cells discarded within that virtual connection by that node since the departure of the previous loss management cell in that virtual connection is stamped at the nodes 24. 26, 28 into one of a multiple of unused Cell Loss Count fields 30.
  • the total cell loss is then calculated at the endpoint of the virtual connection.
  • a combined loss management packet can be used to allow each node to record both accumulated and node-by- node packet loss simultaneously. Other useful information can be recorded in the node-by-node loss measurement management packet, such as the positions of individual discarded packets within a data packet block.
  • the methods described require in several of the embodiments that the management packets experience the same delay and packet loss rate as the user data packets in a particular virtual connection.
  • the simplest way to implement this is to have the management packets follow the same path as the user packets through the switch. In order to accomplish this, the payloads of the management packets must be modified without removing the packets from the data packet stream. Management packets processed in this manner are "associated management packets". Shown in Fig. 6 is a block diagram of an embodiment of a switch with the capability of processing the payloads of associated management packets without removing them from the data packet stream. This embodiment also illustrates one of several possible implementations of local observability of packet delay.
  • a datastream enters the switch 31 at input ports 32.
  • the datastream consists of user data packets 34 interspersed with management packets 36.
  • the user packets 34 and management packets 36 enter the switch input header processor 38, which reads the network routing instructions for each packet and uses them to create an internal routing header that is then attached to that packet. If the packet is a delay management packet, creation of the intemal routing header will include stamping of the time of entry into the internal routing header by the time- of-day processor 40. While it is only necessary to stamp delay management packets with the time of entry, if it is more expedient in terms of processor design, time, etc.. it is not necessary that management packets 36 be distinguished from user data packets 34 by the input header processor 38.
  • all incoming data packets on the input ports 32 can have a time of entry recorded in their intemal routing header.
  • the switch fabric 42 Once the internal routing header has been created, all the data packets are routed through the switch fabric 42 and directed towards their intended output ports 48. Before entering the output ports 48, the packets are routed through the output header processor 44 which removes the intemal routing header from the packets.
  • the processor 44 identifies the delay management packets 36 and uses the time-of-day stamp in the intemal routing header to calculate the delay incurred by each packet in the switch. The time-of-day stamp in the intemal routing header is compared to the current time-of-day.
  • the Delay-stamp processor 44 can merely record the time of exit directly into a timestamp field in the delay management packet 36, allowing the delay calculations to be made later when the packet has reached the virtual connection endpoint.
  • a similar procedure is used in a preferred embodiment to enter packet loss counts into loss management packets. It is not necessary that loss management packets be stamped with a time of entry to the switch by the time-of-day processor 40.
  • a loss management packet is routed through the output header processor 44.
  • the processor identifies it as a loss management packet.
  • the number of packets discarded by the switch for that particular virtual connection since the last loss management packet in that virtual connection was processed by the switch is then written or incremented into the packet loss count field of the packet. This writing can be done either as an additional function of the Delay-stamp processor 46, or by a separate processor operating in a similar manner.
  • the packet loss count is written into the loss management data packet 36 while the previous user data packet 34 in the datastream is being sent to the appropriate output port 48. allowing the loss management data packet to retain its place in the datastream.
  • the time-of-day processor 40 writes the time a delay management packet 36 arrives at the switch directly into a timestamp field in the packet rather than into the intemal routing header.
  • the time-of-day processor 40 will function like the delay-stamp processor 46 in that the time of entry will be recorded in the payload of the delay management packet 36 while the previous user data packet 34 is being sent to the switch fabric 42.
  • the delay-stamp processor 46 then reads the time of entry from the timestamp field of the delay management packet and uses it to calculate the delay by which the processor 46 increments the delay-stamp field in the packet.
  • a switch capable of modifying packets without removing them from their place in the datastream has uses beyond the performance monitoring function.
  • a switch could be utilized to help recovery in the ATM Application Layer by allowing identification of from exactly where in the datastream a cell has been lost.
  • This information in turn can be used to request a retransmittal of that cell only, consuming fewer resources than if an entire data block had to be retransmitted.
  • the same information can be used to minimize the da age caused to the data by the absence of the cell, without retransmittal, via error concealment processes.
  • management packets must for some reason be extracted from the datastream and then diverted to a special processor, as shown in one embodiment in Fig. 7, they do not experience the same delay as user data packets and will not normally retain their positions relative to the user data packets in the same virtual connection.
  • a datastream consisting of user data packets and "associated management packets" 52, interspersed with “non-associated management packets” 54, enters the switch 56 at input ports 50.
  • Associated management packets are those management packets that are intended to be processed in a manner that allows them to retain their place in the datastream, while non-associated management packets are management packets that are processed in such a way that they lose their position with respect to the other packets in the datastream.
  • Non-associated management packets are used to gather cumulative information from each switch, such as the average and maximum packet delays measured in the switch since the previous management packet.
  • the control processor 62 accumulates a set of such performance or usage measure-ments and records this info ⁇ nation into the arriving non-associated management packets 54.
  • the ability to write into the payload of a non-associated type of packet is available in existing switches, such as the GTE Government Systems SPANet switch.
  • the user data packets/associated management packets 52 and some non-associated management packets 54 enter the switch input header processor 58. which reads the network routing instructions for each packet and uses them to create an inte al routing header that is then attached to that packet. If the packet is a non-associated management packet 54, the internal routing header directs the switch fabric 60 to route the packet 54 to the control processor 62 for modification. Once the intemal routing headers have been created by the input header processor 58, the data/associated management packets 52 are routed through the switch fabric 60 and directed through the output header processor 64 towards their intended output ports 68 This means that the non- associated management packet has lost its place in the datastream while it was being modified by the control processor 62.
  • non-associated management packets it is possible to use non-associated management packets to implement performance monitoring functions such as local observability of packet delay and packet losses.
  • This is implemented in one embodiment by having the control processor 62 replace the non-associated management packet 54 that is removed from the datastream for processing with a dummy "marker packet" that holds the place of the non-associated management packet 54 in the datastream.
  • the control processor 62 then separately records the necessary performance monitoring information in the payload of the management packet 54. When it has finished, the control processor 62 replaces the dummy marker packet in the datastream with the appropriate nonassociated management packet 54.
  • Management packets with delay stamping can be used to accurately measure and report both the end-to-end packet delay and the packet delay at each node along a connection, without the need for time-of-day clock synchronization between nodes. Management packets can also accurately measure and report the packet loss at each node along a connection, minimizing the confusion created by misinserted packets.
  • Non-associated management packets allow collection and reporting of cumulative performance and usage information from each node along a connection. The combination of the management packets and switches described allow accurate and detailed QoS measurements along a network connection.

Abstract

Management packets (36, 52, 54) are defined that are modified in the payload by each node (16, 18, 20; 24, 26, 28) along a virtual connection and are used to measure both end-to-end QoS and specific individual intermediate node performance parameters. Management packets (36, 52, 54) are implemented by defining entirely new packets or by modifying ATM OAM cells. Switches or routers for use as intermediate nodes are defined that modify the payload of the management packet (36, 52, 54) and locally measure packet delay and packet loss. An intermediate node measures and records the difference between the arrival and departure times of management packets (36, 52, 54) at that switch (31. 56) utilizing delay-stamp fields within the management packets (36, 52, 54) and either the switch internal routing header or timestamp fields (9) within the packet. At the endpoint of the virtual connection, delay-stamp fields (10) in the management packet (36, 52, 54) indicate individual node delays and the cumulative delay.

Description

PERFORMANCE MONITORING OF AN ATM SWITCH
Technical Field The present invention relates to electronic communications networks and. in particular, to performance monitoring of packet-switched data nerworks. Background Art
The Broadband Integrated Services Digital Network (B-ISDN) can accommodate a variety of packet-oπented protocols, including asynchronous transfer mode (ATM), in which standardized 53-byte data packets are utilized [ITU-T Ree. 1.121. Broadband Aspects of ISDN. Melbourne, Nov. 1988]. ATM and some other packet-oπented protocols are connection-oriented, utilizing what are known in ATM as virtual path connections (VPC) and virtual channel connections (VCC) that stretch between two endpoints across transmission links connected by intermediate nodes. These intermediate nodes are devices that forward and direct the packets to their intended destination endpoint. i.e. routers, or switches. Packet-switched data networks are capable of supporting a simultaneous mix of voice, video, and data services over either constant bit rate (CBR) or variable bit rate (VBR) connections. For each virtual connection, an ATM network can guarantee to the customer a particular level of performance, known as a quality of service (QoS). that is negotiated between the user and the network during connection set-up. Once agreed upon, the QoS can only be changed through renegotiation between the two parties. In general, the deiay experienced by a data packet as it travels along a network connection is an important measure of performance for a packet-switched network. QoS in ATM is specified mainly in terms of acceptable ceil delay and acceptable cell loss rate [ATM Forum, ATM User-Network Interface Specification Version 3. O. Sept. 10. 1992]. ATM supports multiple QoS classes geared to differing application requirements for delay and loss performance. A VCC or VPC is therefore characteπzed by its requested QoS and throughput parameters. A VPC may carry statistically multiplexed VCCs of diffeπng QoS classes, but the overall QoS of the VPC must satisfy the most demanding QoS of the earned VCCs Specified QoS parameter values may be either explicitly requested by the user or implicitly associated with a particular service.
An important concern is management of traffic to ensure that users receive eir guaranteed QoS. The assurance of a particular QoS is a fundamental part of the contractual agreement between the network and the users because certain applications critically depend on receiving a particular QoS from the network. Voice transmissions, for example, are particularly sensitive to end-to-end delay. For this reason, public telephone networks have a maximum allowable delay specification of 600 ms [Chen, Thomas M., Walrand. Jean, and Messerschmitt, David G., "Dynamic Priority Protocols for Packet Voice." IEEE Journal on
Selected Areas in Communications. Vol. 7, No. 5, June 1989, pp. 632-643], with any delay greater than 15 ms generally requiring the use of echo cancellation [McDysan, David E. and Spohn. Darren L.. ATM.' Theory and Application, McGraw-Hill, 1995, p. 355]. Performance management, as a function of network management, has the simultaneous responsibilities of maintaining both the QoS per virtual connection and the efficient utilization of the network. Performance management itself depends largely on the data obtained through performance monitoring, which is conducted through measurement of the QoS per VPC/VCC. Because of the use of statistical multiplexing in ATM networks, the QoS of a virtual connection may in particular be adversely affected by the random flow of traffic on the network. QoS is also degraded by a number of other factors, including network congestion and transmission faults, as well as intermittent error conditions that may cause a gradual deterioration in QoS. Network providers need to be able to detect and diagnose troubles quickly in order to minimize the effect of those troubles on network users.
Performance monitoring can be done either through out-of-service testing or in-service QoS monitoring. Out-of-service testing involves performance measurements made under controlled traffic conditions. This type of testing has a number of drawbacks, not the least of which is that service on the network must be :nterrupted. In addition. Lie results received are only as accurate as the model used to create the controlled traffic conditions. It is entirely possible that the controlled conditions tested may not correctly model some of the conditions actually seen by tne network duπng service. In addition, this method does not allow rapid detection of any subsequent degradation of the network.
In-service QoS monitoring involves continuous performance measurements on actual user traffic. No service interruption occurs and network degradation is instantaneously detected. Continuous performance monitoring is paπicuiarly important in high-speed networks where service deterioration may quickly affect a large amount of user data. In-service QoS measurements are also useful to supplement traffic control algorithms based on assumed traffic models, such as those used in out-of-service testing, that may not accurately characterize the actual traffic.
Current ATM standards provide for the measurement of cell delays and cell loss using Operations and Maintenance (OAM) cells [ITU-T Ree. 1.610, B-ISDN Operations and Maintenance Principles and Functions, Geneva, June 1992; Bellcore. Generic Requirements for Operations of Broadband Switching Systems, TA-NWT00 1248. issue 2. Oct. 1993 ]. OAM cells are specialized versions of standard 53 -byte ATM cells, having a normal 5-byte header with a 48-byte Lπformation field, but carrying management information within the function- specific information fields instead of user data. ATM user cells themselves cannot be used for QoS monitoring because the payload is full of user data and no header fields are available to receive the information that is needed by the performance monitoring system for calculation and storage of cell loss and cell delay. OAM cells are routed by the ATM network in the same manner as user cells, but are distinguished from user cells by the network by the assignment of special values to the VCI and/or PT fields in the 5-byte cell header, utilizing VCI=3. VCI=4. PT=4 or PT=5. OAM cells for performance management are further identified in the information field by a 4-bit OAM Cell Type = 0010. A 4- bit OAM Function Type field in the information field further indicates the specific performance management function performed by the OAM cell, such as forward monitoring (0000), backward reporting (0001), or both (0010). A 10-bit EDC field contains an error detection code that is computed over the information field of the cell. The forward monitoring procedure consists of inserting OAM performance management cells into the cell stream between blocks of user cells at one endpoint of the virtual connection. Backward reporting involves putting observed performance data into an OAM cell at a virtual connection endpoint and then sending the OAM cell backwards in the cell stream direction from which it came, in order to report the performance of that particular virtual connection to the originating endpoint.
To measure cell loss, the OAM cell contains the number of user cells sent in the preceding block (Total User Cell Number) in a 2-byte function-specific field. The destination endpoint of the virtual path counts the number of user cells received between the arrival of the previous OAM cell and the current OAM cell. If fewer cells are received at the destination than are indicated in the current OAM cell, the difference is interpreted to be the number of lost cells.
The major disadvantage of the current network performance monitoring method for measuring cell loss is that it cannot be assured of actually determining the exact number of cells lost. This is because cells can also be misinserted into the cell stream. Therefore, for example, if the OAM cell indicates that N user cells were sent and then N- 1 cells are actually received, it is not possible to be sure whether 1 cell was lost, or whether 2 cells were lost and 1 misinserted, and so on. It is also not possible, using this method, to determine from where along the virtual connection, or from where within a cell block, the cell loss actually occurred. In order to measure cell delay, the OAM performance management cell contains an optional 4-byte Timestamp field that is used to indicate the time that the OAM cell was inserted into the user stream. The timestamp is then compared with the arrival time of the cell at the virtual connection endpoint and the difference is inferred to be the end-to-end delay experienced by a cell along the virtual connection. However, this type of timestamp is meaningful at the destination only if the clocks at the two endpoints of the virtual connection are synchronized to the same time-of-day. something which is generally not practical. In addition, the existing method of delay measurement does not take into account the fact that end-to-end delay actually consists of two distinct types of delay: fixed propagation delay, which is that delay introduced into the connection at a predictable level by the transmission links extending between the network nodes, and random delay. Random delay is that delay which is introduced by each of the intermediate nodes themselves, and can vary widely with network traffic levels and switch buffer capacity. In general, this type of delay is primarily due to queuing delays, which tend to increase with increased traffic load. The existing method of performance monitoring using OAM cells is not able to track the delay that is actually experienced at each node along the virtual connection due to such factors. An alternative way to measure cell delay is to measure round-trip cell delay via an OAM loopback cell, which is identified by OAM Cell Type = 0001 and OAM Function Type = 1000. The loopback cell is sent along the virtual connection and then retumed to its originating endpoint from the destination endpoint or from a prespecified intermediate node. A timestamp allows computation of the round- trip delay back at the originating endpoint, without the need for time-of-day clock synchronization between nodes. The one-way end-to-end delay is then estimated as half the round-trip delay. However, while the loopback cell method can measure the round-trip delays without need for clock synchronization, it gives only an estimate of the one-way cell delay. This estimate may in actuality be very inaccurate if the network congestion or characteristics have altered greatly between the two halves of the OAM cell's trip. Furthermore, this method is also not able to measure the delay occurring at individual intermediate nodes.
All of these disadvantages to current network performance monitoring methods arise generally from the fact that the intermediate nodes in the virtual connection relay the OAM cells but do not actually modify their payloads. Although ATM switches can have the capability to remove and manipulate the header of an ATM cell, it is not normal for a switch to process the entire payload of a cell. When this does occur, it is only done through removal of the cell from its place in the cell stream, which currently prevents such a cell from being used to calculate position dependent parameters such as cell delay and cell loss. For this reason. OAM cells used in performance monitoring are solely a mechanism for measurement of performance between the two endpoints of the virtual connection. The performance monitoring capability of the network is severely limited by this lack of ability of intermediate nodes to add information into the payload of a passing data packet. What has been needed is a way to collect and utilize the information that can be provided to the performance monitoring system by the intermediate nodes. Disclosure of Invention
In one aspect of the invention. a new class of packet is defined which has a payload that is modified by each node along a virtual connection. Management packets are used to accurately measure and report the end-to-end QoS along a virtual connection, as well as to measure and report specific performance parameters at individual intermediate nodes In an ATM embodiment, management cells are implemented by defining an entirely new type of cell or by defining additional fields within existing OAM cells. The switch at each intermediate node modifies the payload of management packets and locally measures delay and packet loss. A function type field specifies the particular performance monitoring function performed by each management packet. The information field of the management packet optionally contains timestamp fields, delay-stamp fields, and packet loss count fields.
In another aspect of the invention. intermediate switches can measure the difference between the arrival time and the departure time of any packet at that switch. For an end-to-end delay accumulating management packet, a delay-stamp field is incremented by the local delay measured at each switch. When the management packet reaches the endpoint of the virtual connection, the delay-stamp field indicates the cumulative amount of time that the packet has been delayed by the nodes of the network.
In another aspect of the invention. a node-by-node delay measurement management packet allows computation and relay of instantaneous delay information by utilizing multiple delay-stamp fields to record each delay encountered on a node by-node basis. At the destination endpoint. the end-to-end delay is computed as the sum of the recorded delay-stamp fields. Alternatively, the switch may record in one timestamp field the time that the packet arrives, and then record the time the packet leaves into another timestamp field, with the calculation of node-by-node delay made when the packet reaches the endpoint of the virtual connection.
According to a further aspect of the invention, intermediate switches can count the number of packets each discards. When a node receives an end-to-end loss accumulating management packet, it increments a packet loss count field by the number of packets discarded by the node within that virtual connection since the previous loss management packet arrived. When the end-to-end loss accumulating management packet arrives at the virtual connection endpoint, the packet loss count field reflects the total number of discarded packets in the preceding block of user packets. Node-by-node loss measurement management packets contain multiple packet loss count fields for recording the number of discarded packets counted at each node since the departure of the previous loss management packet in that virtual connection. The total packet loss is then calculated at the endpoint of the virtual connection.
Associated management packets experience the same delay and packet loss rate as the user data packets in a particular virtual connection, implemented in one embodiment of the invention by having the management packets follow the same path as the user packets through the switch- The pay loads of associated management packets are processed without the packet being removed from the data packet stream. In one embodiment, this is implemented by using the switch internal routing header. When the packets enter the switch input header processor, it uses the network routing instructions for each packet to create an internal routing header that, for a delay management packet, includes the time of entry into the switch. In an alternate embodiment, the time a delay management packet arrives at the switch is recorded directly into a timestamp field in the packet while the previous user data packet is being sent to the switch fabric.
In the former embodiment, all the data packets are routed through the switch fabric and directed towards their intended output ports through an output header processor, which removes the internal routing header from the packets, compares the time-of-day stamp in the intemal routing header of delay management packets to the current time-of-day to calculate the delay incurred by each packet in the switch, and records the difference in the delay-stamp field of the management data packet. This occurs while the previous user data packet in the datastream is being sent to the output port so that the management data packet retains its place in the datastream. A similar procedure is followed if the timestamp field is used, with the output header processor reading the time of entry from that field.
In another aspect of the invention. packet loss counts are recorded in loss management packets. When a loss management packet is routed through the switch output header processor, the processor records into the packet loss count field the number of packets discarded by the switch within that particular virtual connection since the last loss management packet was processed. The packet loss count is recorded while the previous user data packet in the datastream is being sent to the output port, allowing the loss management data packet to retain its place in the datastream. Non-associated management packets are extracted from the datastream and diverted to a special processor, therefore not retaining their positions relative to the user data packets in the same virtual connection. Non-associated management packets are used in one aspect of the invention by a control processor to report cumulative performance or usage measurements. The input header processor creates an internal routing header for non-associated management packets that directs the switch fabric to route the packets to the control processor for modification.
Non-associated management packets are used in another aspect of the invention to implement local measurement of packet delay and packet losses by replacing the non-associated management packet temporarily with a marker packet that holds the place of the non-associated management packet in the datastream while the control processor records the delay or packet loss count information in the payload of the management packet. The marker packet is then replaced in the datastream with the modified non-associated management packet. Local measurement of delay and packet loss is implemented in an alternate embodiment of the invention by routing all data packets through a control processor, recording delay and packet loss count information into appropriate management packets, and performing dummy writes to user data packets. Brief Description of the Drawings Fig. l is a block diagram of an embodiment of the management packet of the present invention for an ATM implementation; Fig. 2 is a block diagram illustrating one embodiment of the method of the present invention, utilizing end-to-end delay accumulating management cells; Fig. 3 is a block diagram illustrating another embodiment of the method of the present invention, utilizing node-by-node delay measurement management cells: Fig. 4 is a block diagram illustrating an embodiment of the method of the present invention, utilizing end-to-end loss accumulating management cells: Fig. 5 is a block diagram illustrating another embodiment of the method of the present invention utilizing node-by-node loss measurement management cells; Fig. 6 is a block diagram illustrating the flow of associated management packets through a switch with local observability of packet delay according to one aspect of the present invention; and Fig. 7 is a block diagram illustrating the flow of non-associated management packets through a switch according to another aspect of the present invention. Best Mode for Carrying Out the Invention
The limitations of prior art network connection performance monitoring methods are eliminated by defining a special class of packet, a "management packet," having a payload that can be modified by each node along a virtual connection. This method can be applied to any packet-switching network. In the preferred ATM embodiment, these "management cells" can optionally be implemented as extensions of current OAM cells. Like OAM cells, management cells carry measurement data along the virtual connection in the cell information field. The major difference between management cells and prior art OAM cells is that the information field of a management cell is modified by all the network nodes along a virtual connection, not just by the virtual connection endpoints. This new capability makes it possible for the performance monitoring system to track the performance of each node along the virtual connection, rather than just the end- to-end performance of the virtual connection as a whole.
Management packets are used to accurately measure the end-to-end QoS experienced by the user packets along any virtual connection. In addition, specific performance parameters measured at individual intermediate nodes are made available to the performance monitoring system. This allows in-service monitoring to be used to accurately identify the QoS actually seen by users and to detect exact causes of performance deterioration before users are seriously affected. The method of the present invention therefore offers vastly improved in-service monitoring of both end-to-end packet delay and packet loss rate. The method requires that the router or switch at each node have specific, unique capabilities, including local processing measurement of packets and local measurement of packet delay and packet loss. Management packets are used to accurately measure end-to-end packet delay, without the need for time-of-day clock synchronization between network connection endpoints, by measuring and reporting the packet delay at each node along a virtual connection. Likewise. management packets allow accurate measurement and reporting of both end-to-end packet loss and packet loss at each node along a virtual connection.
The ATM management cells of the preferred embodiment may be implemented either by defining entirely new cells, or by defining additional fields within existing OAM cells, since existing OAM cells used for performance management have 33 bytes in the information field that are unused. An embodiment of a management cell for an ATM application is shown in Fig. 1. A 53-byte ATM management cell has a standard 5-byte ATM cell header 2. Within this cell header 2, the cell is identified to the network as an OAM cell by specific codes in the VCI and/or PT fields (VCI=3, VCI=4, PT=4 or PT=5). In an alternate embodiment, entirely new VCI and PT codes are defined that uniquely identify the cell to the network as an ATM management cell.
If existing OAM VCI and PT codes are used in the cell header 2, the management cell is instead uniquely identified as a management cell to the network by a special code in the 4-bit OAM Cell Type field 4. A 4-bit Function Type field 6 further indicates the specific performance monitoring function performed by the management cell. As in a regular OAM cell, a 10-bit EDC field 8 contains an error detection code that is computed over the information field of the cell. In the ATM embodiment of Fig. 1 , the Function Type field 6 contains a code that describes to the network the type of performance monitoring function performed by the ATM management cell, labeling the cell as an "end-to-end delay accumulating" cell, a "node-by-node delay measurement" cell, an "end-to-end loss accumulation" cell, a "node-by-node loss measurement" cell, or as some combination of the previous types. Additional codes can be defined for the monitoring of other performance parameters. The information field of the management cell will also optionally contain a 4-byte Timestamp field 9, a Delay- stamp field 10, and a Cell Loss Count field 12, as well as other function-specific fields 14 that are used for additional delay-stamps, cell loss counts, etc., as needed. These fields can also be used for monitoring and reporting other performance parameters.
For measurement of packet delay, a switch is required that can measure the difference between the arrival time and the departure time of any packet at that switch, a capability known as "local observability" of packet delay. Because most switches have at least some buffering capability, the delay incurred by a packet in the switch can be highly variable, particularly where a priority-based buffering scheme is implemented. Packet delay within a switch tends to increase as network traffic increases. Existing packet switches are not currently implemented so as to have the ability to measure the delay experienced by a packet within the switch itself
The management packet method utilizes measurements of the individual packet delay experienced between the input and output of each of the intermediate switches. Existing switches used with performance monitoring OAM cells have the ability to timestamp OAM cells when they enter a cell stream. In the management packet method, this activity can be replicated at each intermediate node. If the management packet is an end-to-end delay accumulating management cell, the intermediate node takes the Delay-stamp field 10. which was set to zero at the originating endpoint of the virtual connection, and increments it by the local delay measured at that switch, as illustrated in Fig. 2 for an ATM embodiment.
In Fig. 2, the Delay-stamp field 10 initially contains a delay TO from previous intermediate nodes. At node 16, the time the cell of Fig. 1 enters the node 16 is recorded in the Timestamp field 9. This field is then compared to the time the cell leaves the node and the Delay-stamp field 10 is incremented by the difference. Tl . At node 2 18, a new time in is recorded in the Timestamp field 9, corresponding to the time the cell has entered the node 18. This field is again compared to the time the cell exits the node 18 and the difference, T2 is used to increment the value in the Delaystamp field 10. A similar procedure is followed at node 3 20. and so on, until the endpoint of the virtual connection is reached. Alternatively, the time a packet enters each node can be incorporated into a field in -13-
the inte al routing header used by that node to direct the packet to the proper output, making use of a timestamp field within the packet unnecessary. The switch then examines the time of entry recorded in the routing header to calculate the delay that is written or incremented into the Delay-stamp field. Following either procedure, when the management cell of Fig. 2 finally reaches the endpoint of the virtual connection, the Delay-stamp field 10 indicates the cumulative amount of time that the cell has been variably delayed by the nodes of the network. If desired, this delay can then be added to the value of fixed propagation delay caused by the transmission links between the nodes in order to arrive at the total time that the cell has spent in the network. While this delay- stamp approach requires the new switch capability of local observability of packet delay, it has a big advantage over the OAM cell performance monitoring method in that it does not require any time-of-day synchronization between the switches in a network. For monitoring of packet delay in more detail, a node-by-node delay measurement management packet may be used. A particular advantage of the node- by-node delay measurement management packet is that it allows the computation and relay of instantaneous delay information. This type of management packet contains multiple delay-stamp fields for recording each local delay encountered on a node-by-node basis, as illustrated in Fig. 3 for an ATM embodiment. In Fig. 3. at each node the cell is stamped in the Timestamp field O, or in the separate internal routing header, with the time of entry into the node. This time is then compared to the time of exit from that node and the calculated delay is recorded into the cell in one of a multiple of unused Delay-stamp fields 22 This allows monitoring of the delay performance of each intermediate node at the virtual connection endpoint. If desired, the end-to-end cell delay can also be computed at the destination endpoint, being the sum of the delay-stamp fields. Alternatively, additional timestamp fields can be used for each node, allowing the switch to merely record in one timestamp field the time that the packet arrives, and then record the time the packet leaves into another timestamp field. The calculation of node-by-node and end-to-end delay can then be made when the packet has reached the endpoint of the virtual connection. A further alternative is for the cell to be a combination delay cell, allowing the switch to record both the accumulated and the node-by-node delays in the same packet. Measurement of packet loss requires a switch with local observability of packet loss, allowing the switch to count the number of packets it discards per virtual connection. Packet discards occur for various reasons, such as switch buffer overflow due to excessive traffic, corrupted routing information in a cell header field or on the
Figure imgf000016_0001
connection, and when packets have failed to arrive at their destination within a pre-specified time and have therefore "expired." Each switch must count the number of packets it discards within each virtual connection. This capability is already available in certain existing switches, such as the AT&T GlobeView-2000 switch [See AT&T, Globe View-2000 Broadband System Network: Network Management, 255-179-079, Issue 1.0, Release 1.3, 1995] and the Newbridge 36150 MainStreet switch [See Newbridge, 36150 MainStreet
ATMnet Access Switch Technical Practices, NNP 95-1602-01 -00-A, Release 3.1, Generic A 11113. 1995 ].
The management packet used to measure packet loss can be the same packet as used to measure packet delay or a different packet. Packet loss is measured according to one embodiment of the present invention through the insertion of end-to-end loss accumulating management packets into a virtual connection between blocks of user packets, as illustrated in Fig. 4 for an ATM embodiment. Each end-to-end loss accumulating management cell contains a Cell Loss Count field 12 that is initialized to zero. When node 24 receives the end-to- end loss accumulating management cell with a Cell Loss Count field value NO, it increments the Cell Loss Count field 12 by the number of cells discarded by the node 24 in that virtual connection since the previous loss management cell arrived in that particular virtual connection. The cell is then received by node 226, which increments the Cell Loss Count field 12 by the number of cells discarded within that virtual connection by that node. A similar procedure is followed at the next node. node 3 28. and so on. By the time the end-to-end loss accumulating management cell ultimately arrives at its destination virtual connection endpoint. the Cell Loss Count field 12 accurately reflects the total number of discarded cells in the preceding block of user cells. In contrast to the use of existing OAM performance management cells, it is not required that a node know how many cells to expect in the user cell block, since each node accurately counts how many cells it discards.
For monitoring of packet loss in greater detail, node-by-node loss measurement management packets containing multiple fields to record the number of discarded packets counted at each node may be used. An illustration of this is shown in Fig. 5 for an ATM embodiment. At each node, the number of cells discarded within that virtual connection by that node since the departure of the previous loss management cell in that virtual connection is stamped at the nodes 24. 26, 28 into one of a multiple of unused Cell Loss Count fields 30. Just as for the node-by-node delay measurement management cell, the total cell loss is then calculated at the endpoint of the virtual connection. A combined loss management packet can be used to allow each node to record both accumulated and node-by- node packet loss simultaneously. Other useful information can be recorded in the node-by-node loss measurement management packet, such as the positions of individual discarded packets within a data packet block.
The methods described require in several of the embodiments that the management packets experience the same delay and packet loss rate as the user data packets in a particular virtual connection. The simplest way to implement this is to have the management packets follow the same path as the user packets through the switch. In order to accomplish this, the payloads of the management packets must be modified without removing the packets from the data packet stream. Management packets processed in this manner are "associated management packets". Shown in Fig. 6 is a block diagram of an embodiment of a switch with the capability of processing the payloads of associated management packets without removing them from the data packet stream. This embodiment also illustrates one of several possible implementations of local observability of packet delay.
In Fig. 6, a datastream enters the switch 31 at input ports 32. The datastream consists of user data packets 34 interspersed with management packets 36. The user packets 34 and management packets 36 enter the switch input header processor 38, which reads the network routing instructions for each packet and uses them to create an internal routing header that is then attached to that packet. If the packet is a delay management packet, creation of the intemal routing header will include stamping of the time of entry into the internal routing header by the time- of-day processor 40. While it is only necessary to stamp delay management packets with the time of entry, if it is more expedient in terms of processor design, time, etc.. it is not necessary that management packets 36 be distinguished from user data packets 34 by the input header processor 38. Instead, all incoming data packets on the input ports 32 can have a time of entry recorded in their intemal routing header. Once the internal routing header has been created, all the data packets are routed through the switch fabric 42 and directed towards their intended output ports 48. Before entering the output ports 48, the packets are routed through the output header processor 44 which removes the intemal routing header from the packets. At this time, the processor 44 identifies the delay management packets 36 and uses the time-of-day stamp in the intemal routing header to calculate the delay incurred by each packet in the switch. The time-of-day stamp in the intemal routing header is compared to the current time-of-day. and the difference is written or incremented by the Delay-stamp processor 44 into the delay-stamp field of the management data packet 36, while the previous user data packet 34 in the datastream is being sent to the appropriate output port 48. In this way, the management data packet retains its place in the datastream. In an alternate embodiment, the delay-stamp processor 46 can merely record the time of exit directly into a timestamp field in the delay management packet 36, allowing the delay calculations to be made later when the packet has reached the virtual connection endpoint. A similar procedure is used in a preferred embodiment to enter packet loss counts into loss management packets. It is not necessary that loss management packets be stamped with a time of entry to the switch by the time-of-day processor 40. but this may be done if it is more expedient to do so. When a loss management packet is routed through the output header processor 44. the processor identifies it as a loss management packet. The number of packets discarded by the switch for that particular virtual connection since the last loss management packet in that virtual connection was processed by the switch is then written or incremented into the packet loss count field of the packet. This writing can be done either as an additional function of the Delay-stamp processor 46, or by a separate processor operating in a similar manner. The packet loss count is written into the loss management data packet 36 while the previous user data packet 34 in the datastream is being sent to the appropriate output port 48. allowing the loss management data packet to retain its place in the datastream. In an alternate embodiment, the time-of-day processor 40 writes the time a delay management packet 36 arrives at the switch directly into a timestamp field in the packet rather than into the intemal routing header. In this case, the time-of-day processor 40 will function like the delay-stamp processor 46 in that the time of entry will be recorded in the payload of the delay management packet 36 while the previous user data packet 34 is being sent to the switch fabric 42. The delay-stamp processor 46 then reads the time of entry from the timestamp field of the delay management packet and uses it to calculate the delay by which the processor 46 increments the delay-stamp field in the packet.
A switch capable of modifying packets without removing them from their place in the datastream has uses beyond the performance monitoring function. For example, such a switch could be utilized to help recovery in the ATM Application Layer by allowing identification of from exactly where in the datastream a cell has been lost. This information in turn can be used to request a retransmittal of that cell only, consuming fewer resources than if an entire data block had to be retransmitted. Altematively, the same information can be used to minimize the da age caused to the data by the absence of the cell, without retransmittal, via error concealment processes.
If management packets must for some reason be extracted from the datastream and then diverted to a special processor, as shown in one embodiment in Fig. 7, they do not experience the same delay as user data packets and will not normally retain their positions relative to the user data packets in the same virtual connection. In Fig. 7. a datastream consisting of user data packets and "associated management packets" 52, interspersed with "non-associated management packets" 54, enters the switch 56 at input ports 50. Associated management packets are those management packets that are intended to be processed in a manner that allows them to retain their place in the datastream, while non-associated management packets are management packets that are processed in such a way that they lose their position with respect to the other packets in the datastream. Non-associated management packets are used to gather cumulative information from each switch, such as the average and maximum packet delays measured in the switch since the previous management packet. The control processor 62 accumulates a set of such performance or usage measure-ments and records this infoπnation into the arriving non-associated management packets 54. The ability to write into the payload of a non-associated type of packet is available in existing switches, such as the GTE Government Systems SPANet switch.
In the embodiment of Fig. 7, the user data packets/associated management packets 52 and some non-associated management packets 54 enter the switch input header processor 58. which reads the network routing instructions for each packet and uses them to create an inte al routing header that is then attached to that packet. If the packet is a non-associated management packet 54, the internal routing header directs the switch fabric 60 to route the packet 54 to the control processor 62 for modification. Once the intemal routing headers have been created by the input header processor 58, the data/associated management packets 52 are routed through the switch fabric 60 and directed through the output header processor 64 towards their intended output ports 68 This means that the non- associated management packet has lost its place in the datastream while it was being modified by the control processor 62.
It is possible to use non-associated management packets to implement performance monitoring functions such as local observability of packet delay and packet losses. This is implemented in one embodiment by having the control processor 62 replace the non-associated management packet 54 that is removed from the datastream for processing with a dummy "marker packet" that holds the place of the non-associated management packet 54 in the datastream. The control processor 62 then separately records the necessary performance monitoring information in the payload of the management packet 54. When it has finished, the control processor 62 replaces the dummy marker packet in the datastream with the appropriate nonassociated management packet 54.
Altematively, local observability of such performance parameters as delay and packet loss can be implemented using an alternate version of the control processor. In this embodi-ment, all data packets are routed through the control processor rather than just non-associated management packets. Parameters such as delay and packet loss count information are recorded into the payloads of appropriate management packets, while dummy writes are made to the payloads of user data packets, leaving the user data unchanged. Although this embodi-iment potentially requires some unnecessary write operations, it maintains the order of all packets in the datastream, and may provide efficiencies in implementation or in other areas.
Use of management packets provides an improvement over current OAM performance monitoring methods because intermediate nodes along a connection can add their own local performance data into the payload of the packets. Management packets with delay stamping can be used to accurately measure and report both the end-to-end packet delay and the packet delay at each node along a connection, without the need for time-of-day clock synchronization between nodes. Management packets can also accurately measure and report the packet loss at each node along a connection, minimizing the confusion created by misinserted packets. Non-associated management packets allow collection and reporting of cumulative performance and usage information from each node along a connection. The combination of the management packets and switches described allow accurate and detailed QoS measurements along a network connection.
Modifications and substitutions by one of ordinary skill in the an are considered to be within the scope of the present invention. which is not to be limited except by the claims which follow.

Claims

Claims:
1. A method for performance monitoring in an electronic communications network having at least one network connection, each of said network connections having a network connection originating endpoint. a network connection destination endpoint, and at least one network connection intermediate node, comprising, in combination, the steps of: creating at least one management data packet, each of said management data packets comprising, in combination: routing header means; packet function field; and at least one local performance monitoring field; inserting at least one of said management data packets into a stream of user data packets at one of said network connection originating endpoints; measuring at least one performance monitoring parameter at each of said network connection intermediate nodes; for each of said management data packets, recording, at each of said network connection intermediate nodes, in one of said local performance monitoring fields in said management data packet, said performance monitoring parameter measured at said network connection intermediate node; and extracting the values in said local performance monitoring fields after said management data packet reaches said network connection destination endpoint.
2. The method of claim 1, further comprising the steps, in combination, of: including in said local performance monitoring fields at least one delaystamp field, initialized to a predetermined value; detecting entry of each of said management data packets into any of said network connection intermediate nodes; for each of said management data packets and said network connection intermediate nodes, recording the time of entry of said management data packet into said network connection intermediate node; detecting the imminent exit of said management data packet from said network connection intermediate node; comparing the time of said imminent exit of said management data packet to said recorded time of entry; and incrementing the value in one of said delay-stamp fields by the difference between said time of entry and said time of exit.
3. The method of claim 2, further including the step of adding the values in all said incremented ones of said delay-stamp fields of said management data packet, in order to get a total accumulated delay value, when said management data packet reaches said network connection destination endpoint.
4. The method of claim 1 , wherein each of said management data packets further include a packet type field, identifying to said network connection intermediate nodes and said network connection destination endpoint that said data packet is a management data packet.
5. The method of claim 1 , wherein said routing header means of each of said management data packets contains a field identifying to said network connection intermediate nodes and said network connection destination endpoint that said data packet is a management data packet.
6. The method of claim 2, further including the steps, in combination, of: including in said local performance monitoring fields at least one timestamp field; and for each of said delay management packets, recording, in one of said timestamp fields, the time of entry of said management packet into said network connection intermediate node.
7. The method of claim 6, further including the step of adding the values in all said incremented ones of said delay-stamp fields of said management packet, in order to get a total accumulated delay value, when said management packet reaches said network connection destination endpoint.
8. The method of claim 6, further including the step of recording said time of imminent exit in an unused one of said timestamp fields of said delay management packet.
9. The method of claim 8, further comprising the step of comparing each of said timestamp fields containing said times of entry of said delay management packet into each of said network connection intermediate nodes to the corresponding one of said timestamp fields containing said time of imminent exit from said network connection intermediate node in order to obtain a delay value for each of said network connection intermediate nodes.
10. The method of claim 2, wherein said time of entry of said management packet into said network connection intermediate node is recorded in an intemal routing header created at said network connection intermediate node and attached to said management packet while in said network connection intermediate node.
1 1. The method of claim 10, further including the step of adding the values in all said incremented ones of said delay-stamp fields of said management packet. in order to get a total accumulated delay value, when said management packet reaches said network connection destination endpoint.
12. The method of claim 1, further comprising the steps, in combination, of: including in said local performance monitoring fields at least one discarded packet count field, initialized to a predetermined value; counting at each of said network connection intermediate nodes the number of said user data packets that are discarded by said network connection intermediate node between the departure of an immediately previous one of said management data packets and the departure of a current one of said management data packets; and incrementing the value in one of said discarded packet count fields of said current one of said management data packets by said discarded number of said user data packets.
13. The method of claim 12, further comprising the step of adding the values in all said incremented ones of said discarded packet count fields of said management data packet in order to get a total accumulated discarded user packet value, when said management data packet reaches said network connection destination endpoint.
14. The method of claim 2, further comprising the steps, in combination, of: including in said local performance monitoring fields at least one discarded packet count field, initialized to a predetermined value; counting at each of said network connection intermediate nodes the number of said user data packets that are discarded by said network connection intermediate node between the departure of an immediately previous one of said management data packets and the departure of a current one of said management data packets; and incrementing the value in one of said discarded packet count fields of said current one of said management data packets by said discarded number of said user data packets.
15. The method of claim 14, further comprising the step of adding the values in all said incremented ones of said discarded packet count fields of said management data packet, in order to get a total accumulated discarded user packet value, when said management data packet reaches said network connection destination endpoint.
16. The method of claim 15, further comprising the step of adding the values in all said incremented ones of said delay-stamp fields of said management data packet, in order to get a total accumulated delay value, when said management data packet reaches said network connection destination endpoint.
17. A router for use in an electronic communications network comprising, in combination: at least one input for receiving data packets from at least one network connection; connections; at least one output for relaying data packets to one of said network means for detection of entry of a management packet into said router; means for associatively recording the time of entry of said management packet into said router; means for detection of the imminent exit of said management packet from said router; and means for recording a parameter in said management packet that can be used to obtain the delay incurred by said management packet in said router.
18. The router of claim 17 , further comprising means for calculation of the delay incurred by said data packet in said router by utilizing said recorded time of entry and said time of inirninent exit and wherein said means for recording a parameter records said delay in said management packet.
19. The router of claim 18 wherein said means for associatively recording said time of entry records said time of entry in an inte al routing header created by said network connection intermediate node and attached to said management packet while in said router.
20. The router of claim 17 wherein said means for associatively recording said time of entry records said time of entry in a timestamp field in said management packet.
21. The router of claim 20. further including means for recording said time of imminent exit of said management packet from said router into said management packet.
22. A performance monitored switching mechanism for concurrently routing each of a plurality of data packets in a datastream, each of said data packets comprising a header and a payload. to a selected one of plural output data pathways in accordance with a destination address, said mechanism comprising, in combination: a plurality of input ports for concurrently receiving said data packets; a plurality of output data pathways; switching means connected between said input ports and said output data pathways; and means for modifying said payload of one of said data packets without said modified data packet losing its position in said datastream of said data packets.
23. The switching mechanism of claim 22, wherein said means for modifying said payload of one of said data packets further comprises, in combination: means for extracting said data packet from said datastream; means for replacing said extracted data packet with a marker packet; means for modifying said payload of said extracted data packet; and means for replacing said marker packet in said datastream with said modified extracted data packet.
24. The switching mechanism of claim 22, further comprising in combination: means for detection of entry of one of said data packets into said switching mechanism; means for associatively recording the time of entry of said data packet into said switching mechanism; means for detection of the imminent exit of said data packet from said switching mechanism; means for calculating, utilizing said recorded time of entry, the delay incurred by said data packet in said switching mechanism; and means for recording said delay in said payload of said data packet.
25. The switching mechanism of claim 22 wherein said means for modifying said payload of one of said data packets further comprises, in combination: means for selecting one of said data packets in said datastream to be modified in said payload; means for writing data to said payload of said selected means for performing dummy writes to non-selected ones of said data packets in said datastream.
PCT/US1997/004999 1996-04-01 1997-03-27 Performance monitoring of an atm switch WO1997037310A1 (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
JP53541997A JP3894232B2 (en) 1996-04-01 1997-03-27 ATM switch performance monitoring
CA002250278A CA2250278C (en) 1996-04-01 1997-03-27 Performance monitoring of an atm switch
EP97917052A EP0892958A4 (en) 1996-04-01 1997-03-27 Performance monitoring of an atm switch

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US08/625,102 US5793976A (en) 1996-04-01 1996-04-01 Method and apparatus for performance monitoring in electronic communications networks
US08/625,102 1996-04-01

Publications (1)

Publication Number Publication Date
WO1997037310A1 true WO1997037310A1 (en) 1997-10-09

Family

ID=24504598

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1997/004999 WO1997037310A1 (en) 1996-04-01 1997-03-27 Performance monitoring of an atm switch

Country Status (5)

Country Link
US (1) US5793976A (en)
EP (1) EP0892958A4 (en)
JP (1) JP3894232B2 (en)
CA (1) CA2250278C (en)
WO (1) WO1997037310A1 (en)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2000024223A1 (en) * 1998-10-21 2000-04-27 Celogic Method for data qualification and for assisting in fault localization on a digital telecommunication network
FR2785115A1 (en) * 1998-10-21 2000-04-28 Celogic Client server communications network link validation/qualification/fault detection having fault detection discriminating anomalies and analysis finding anomaly/fault origin/nature.
GB2381696B (en) * 2001-10-24 2004-10-13 Zarlink Semiconductor Vn Inc Methods of performance estimation in provisioning delay intolerant data services
EP2211508A1 (en) * 2009-01-22 2010-07-28 IBBT vzw Method and device for characterising a data flow in a network for transferring media data
GB2476116A (en) * 2009-12-14 2011-06-15 Dublin Inst Of Technology Real-time VoIP transmission quality predictor and quality-driven de jitter buffer
EP2381622A1 (en) * 2010-04-23 2011-10-26 Alcatel Lucent Update of a cumulative residence time of a packet in a packet-switched communication network
EP2458787A4 (en) * 2009-11-27 2012-05-30 Huawei Tech Co Ltd Method, device and system for measuring asymmetric time delay of communication paths
EP2562963A1 (en) * 2011-08-25 2013-02-27 Honeywell International Inc. Embedded end-to-end delay information for data networks
EP2611069A1 (en) * 2011-12-29 2013-07-03 British Telecommunications Public Limited Company Distributed system management
WO2013098541A1 (en) 2011-12-29 2013-07-04 British Telecommunications Plc Distributed system management
EP2637370A1 (en) * 2012-02-24 2013-09-11 British Telecommunications public limited company Distributed system management
EP2648365A1 (en) * 2012-02-24 2013-10-09 British Telecommunications public limited company Distributed system management
EP2648099A1 (en) * 2012-02-24 2013-10-09 British Telecommunications public limited company Distributed system management
US8964755B2 (en) 2011-12-30 2015-02-24 British Telecommunications Public Limited Company Obtaining information from data items

Families Citing this family (200)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE69330904T2 (en) * 1992-12-04 2002-06-20 At & T Corp Packet network interface
DE69636158T2 (en) * 1996-01-29 2006-09-28 Agilent Technologies, Inc. (n.d.Ges.d.Staates Delaware), Palo Alto A method and arrangement for performing quality of service measurements on a connection over a network
JP3382769B2 (en) * 1996-02-15 2003-03-04 日本電気株式会社 Transmission equipment with performance monitoring function
JP3175757B2 (en) * 1996-08-13 2001-06-11 日本電気株式会社 Debug system
US6581104B1 (en) * 1996-10-01 2003-06-17 International Business Machines Corporation Load balancing in a distributed computer enterprise environment
US6421731B1 (en) * 1996-10-29 2002-07-16 Telxon Corporation Dynamic next hop routing protocol
US6188690B1 (en) * 1996-12-12 2001-02-13 Pmc-Sierra, Inc. Method and apparatus for high speed, scalable communication system
US5974237A (en) * 1996-12-18 1999-10-26 Northern Telecom Limited Communications network monitoring
JP3497338B2 (en) * 1997-01-08 2004-02-16 株式会社日立製作所 Network system with distributed log batch management function
US5991824A (en) * 1997-02-06 1999-11-23 Silicon Graphics, Inc. Method and system for simultaneous high bandwidth input output
US6449259B1 (en) * 1997-03-31 2002-09-10 Lucent Technologies Inc. Communication controller
US6366581B1 (en) * 1997-04-02 2002-04-02 Fujitsu Network Communications, Inc. Method and apparatus for generating permanent virtual connections using graphical user interface
US6052383A (en) * 1997-05-29 2000-04-18 3Com Corporation LAN to ATM backbone switch module
US6125105A (en) * 1997-06-05 2000-09-26 Nortel Networks Corporation Method and apparatus for forecasting future values of a time series
US7260060B1 (en) * 1997-06-07 2007-08-21 Nortel Networks Limited Call admission control
US6154448A (en) * 1997-06-20 2000-11-28 Telefonaktiebolaget Lm Ericsson (Publ) Next hop loopback
US6052726A (en) * 1997-06-30 2000-04-18 Mci Communications Corp. Delay calculation for a frame relay network
US6411998B1 (en) * 1997-09-08 2002-06-25 International Business Machines Corporation World wide web internet delay monitor
US6041352A (en) * 1998-01-23 2000-03-21 Hewlett-Packard Company Response time measuring system and method for determining and isolating time delays within a network
US6188674B1 (en) * 1998-02-17 2001-02-13 Xiaoqiang Chen Method and apparatus for packet loss measurement in packet networks
US6092108A (en) * 1998-03-19 2000-07-18 Diplacido; Bruno Dynamic threshold packet filtering of application processor frames
SE513188C2 (en) * 1998-03-25 2000-07-24 Ericsson Telefon Ab L M Quota algorithm for monitoring disturbance processes
DE19837245A1 (en) * 1998-08-17 2000-02-24 Siemens Ag Error detection method for ATM communications system
US7707600B1 (en) * 1998-08-21 2010-04-27 Intel Corporation Confirming video transmissions
EP0986226B1 (en) * 1998-09-11 2007-01-17 Hitachi, Ltd. Ip packet communication apparatus
US6594284B1 (en) * 1998-09-16 2003-07-15 Cirrus Logic, Inc. Network synchronization
US6683856B1 (en) * 1998-10-09 2004-01-27 Lucent Technologies Inc. Method and apparatus for measuring network performance and stress analysis
US6360270B1 (en) * 1998-11-16 2002-03-19 Hewlett-Packard Company Hybrid and predictive admission control strategies for a server
US6545979B1 (en) * 1998-11-27 2003-04-08 Alcatel Canada Inc. Round trip delay measurement
JP3230671B2 (en) * 1999-01-14 2001-11-19 日本電気株式会社 Packet billing device
GB2348775B (en) * 1999-03-01 2003-05-28 Motorola Ireland Ltd Analysing a packet radio cellular communications network
US6711137B1 (en) 1999-03-12 2004-03-23 International Business Machines Corporation System and method for analyzing and tuning a communications network
US7167860B1 (en) 1999-03-25 2007-01-23 Nortel Networks Limited Fault tolerance for network accounting architecture
US20020091636A1 (en) * 1999-03-25 2002-07-11 Nortel Networks Corporation Capturing quality of service
GB9909436D0 (en) * 1999-04-23 1999-06-23 Pact Routing device
US6647413B1 (en) * 1999-05-28 2003-11-11 Extreme Networks Method and apparatus for measuring performance in packet-switched networks
JP3549774B2 (en) * 1999-06-01 2004-08-04 富士通株式会社 Network interconnection device and network interconnection method
CN1363170A (en) * 1999-06-24 2002-08-07 西门子公司 Protocol device for a protocol system for transmitting messages
US6437783B1 (en) * 1999-09-13 2002-08-20 Intel Corporation Method and system for simultaneously displaying the throughput on multiple busses
US6674720B1 (en) * 1999-09-29 2004-01-06 Silicon Graphics, Inc. Age-based network arbitration system and method
GB2355901B (en) * 1999-11-01 2003-10-01 Mitel Corp Marker packet system and method for measuring audio network delays
US7523190B1 (en) * 1999-12-23 2009-04-21 Bickerstaff Cynthia L Real-time performance assessment of large area network user experience
JP3376981B2 (en) * 1999-12-24 2003-02-17 日本電気株式会社 Connection control equipment, physical layer LSI and connection control system
US7990882B1 (en) 1999-12-30 2011-08-02 Avaya Inc. Adaptively maintaining quality of service (QoS) in distributed PBX networks
EP1447998A1 (en) * 1999-12-30 2004-08-18 Nortel Networks Limited Adaptively maintaining quality of service (Qos) in distributed PBX networks
US7145881B1 (en) * 2000-04-10 2006-12-05 Peregrine Systems, Inc. Method of determining the route of packets through a network of communicating objects
US7016309B1 (en) * 2000-04-12 2006-03-21 Cisco Technology, Inc. Method and apparatus for calculating packet loss for a communication circuit
JP2001313670A (en) * 2000-04-28 2001-11-09 Oki Electric Ind Co Ltd Method for managing network
US6975617B2 (en) * 2000-05-03 2005-12-13 Agilent Technologies, Inc. Network monitoring system with built-in monitoring data gathering
WO2001091375A2 (en) * 2000-05-19 2001-11-29 Williams Communications Llc Method and system for measuring one-way delay variation
FR2809199B1 (en) * 2000-05-22 2003-08-01 Jean Pierre Christie METHOD FOR PERFORMING PERFORMANCE TESTS OF ACCESSIBLE COMPUTER EQUIPMENT VIA A TELECOMMUNICATION NETWORK
US6845100B1 (en) * 2000-08-28 2005-01-18 Nokia Mobile Phones Ltd. Basic QoS mechanisms for wireless transmission of IP traffic
US6915348B1 (en) * 2000-09-14 2005-07-05 Nortel Networks Limited Validation of a connection between arbitrary end-points in a communications network using an augmented SPE
KR100333394B1 (en) * 2000-09-18 2002-04-25 구자홍 Apparatus for FAX transfer result checking in FAX transfer system by using IP network
US7406539B2 (en) 2000-10-17 2008-07-29 Avaya Technology Corp. Method and apparatus for performance and cost optimization in an internetwork
US8023421B2 (en) 2002-07-25 2011-09-20 Avaya Inc. Method and apparatus for the assessment and optimization of network traffic
US7487237B2 (en) * 2000-10-17 2009-02-03 Avaya Technology Corp. Load optimization
US7336613B2 (en) * 2000-10-17 2008-02-26 Avaya Technology Corp. Method and apparatus for the assessment and optimization of network traffic
US7349994B2 (en) 2000-10-17 2008-03-25 Avaya Technology Corp. Method and apparatus for coordinating routing parameters via a back-channel communication medium
US7720959B2 (en) 2000-10-17 2010-05-18 Avaya Inc. Method and apparatus for characterizing the quality of a network path
US7756032B2 (en) * 2000-10-17 2010-07-13 Avaya Inc. Method and apparatus for communicating data within measurement traffic
AU2002213287A1 (en) 2000-10-17 2002-04-29 Routescience Technologies Inc Method and apparatus for performance and cost optimization in an internetwork
WO2002033893A2 (en) * 2000-10-17 2002-04-25 Routescience Technologies, Inc. Method and apparatus for communicating data within measurement traffic
US7080161B2 (en) * 2000-10-17 2006-07-18 Avaya Technology Corp. Routing information exchange
US7363367B2 (en) * 2000-10-17 2008-04-22 Avaya Technology Corp. Systems and methods for robust, real-time measurement of network performance
EP1199839A1 (en) * 2000-10-19 2002-04-24 THOMSON multimedia Method for making bridge aware nodes communicate over hiperlan 2 bridges
US7035217B1 (en) * 2000-10-20 2006-04-25 Cisco Technology, Inc. Router-assisted multicast congestion control
EP1206067A1 (en) * 2000-11-06 2002-05-15 Agilent Technologies, Inc. (a Delaware corporation) Method of and apparatus for network measurement
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
US6847657B1 (en) 2000-12-22 2005-01-25 Applied Micro Circuits Corporation System and method for programming synchronization criteria in a multidimensional digital frame structure
US6836485B1 (en) 2000-12-22 2004-12-28 Applied Micro Circuits Corporation System and method for programming loss of synchronization in a multidimensional digital frame structure
US6931006B1 (en) 2000-12-22 2005-08-16 Applied Micro Circuits Corporation System and method for selectively broadcasting a multidimensional digital frame structure
US7158535B1 (en) 2000-12-22 2007-01-02 Applied Micro Circuits Corporation System and method for translating overhead bytes in a multidimensional digital frame structure
US6973099B1 (en) 2000-12-22 2005-12-06 Applied Micro Circuits Corporation System and method for programming the value of frame synchronization words in a multidimensional digital frame structure
US6961350B1 (en) 2000-12-22 2005-11-01 Applied Micro Circuits Corporation System and method for segmenting a multidimensional digital frame structure
US6965618B1 (en) 2000-12-22 2005-11-15 Applied Micro Circuits Corporation System and method for programming the bit error rate of frame synchronization words in a multidimensional digital frame structure
US7054336B1 (en) 2000-12-22 2006-05-30 Applied Micro Circuits Corporation System and method for programming the quantity of frame synchronization words in a multidimensional digital frame structure
US6684351B1 (en) 2000-12-22 2004-01-27 Applied Micro Circuits Corporation System and method for diagnosing errors in multidimensional digital frame structure communications
US7023881B1 (en) 2000-12-22 2006-04-04 Applied Micro Circuits Corporation System and method for selectively scrambling multidimensional digital frame structure communications
US6973100B1 (en) 2000-12-22 2005-12-06 Applied Micro Circuits Corporation System and method for programming the location of frame synchronization words in a multidimensional digital frame structure
US6715113B1 (en) 2000-12-22 2004-03-30 Applied Micro Circuits Corporation Feedback system and method for optimizing the reception of multidimensional digital frame structure communications
US6873605B1 (en) 2001-01-02 2005-03-29 Applied Micro Circuits Corporation Bidirectional line switch ring system and method
US6961366B1 (en) 2001-01-02 2005-11-01 Applied Micro Circuits Corporation System and method for redundant path connections in digital communications network
US6882644B1 (en) 2001-01-02 2005-04-19 Applied Micro Circuits Corporation System and method for diagnostic multicast switching
US7050400B1 (en) 2001-03-07 2006-05-23 At&T Corp. End-to-end connection packet loss detection algorithm using power level deviation
US20020184363A1 (en) * 2001-04-20 2002-12-05 Steven Viavant Techniques for server-controlled measurement of client-side performance
US6937594B2 (en) * 2001-04-27 2005-08-30 Lucent Technologies Inc. Loop back testing for multi-protocol hybrid networks
US6944460B2 (en) * 2001-06-07 2005-09-13 Telefonaktiebolaget L M Ericsson (Publ) System and method for link adaptation in communication systems
ATE356518T1 (en) * 2001-08-09 2007-03-15 Ascom Schweiz Ag ANALYSIS OF A DATA TRANSMISSION SYSTEM
US7072291B1 (en) 2001-08-23 2006-07-04 Cisco Technology, Inc. Devices, softwares and methods for redundantly encoding a data stream for network transmission with adjustable redundant-coding delay
US6721686B2 (en) * 2001-10-10 2004-04-13 Redline Networks, Inc. Server load testing and measurement system
WO2003036889A1 (en) * 2001-10-25 2003-05-01 Worldcom, Inc. Communication session quality indicator
GB2382502B (en) * 2001-11-23 2005-10-19 Actix Ltd Network testing systems
ATE324717T1 (en) * 2001-11-28 2006-05-15 Bridgeco Ag METHOD FOR SYNCHRONIZATION IN NETWORKS
US6999417B1 (en) * 2001-11-30 2006-02-14 Cisco Technology, Inc. Devices, softwares and methods for incorporating burstiness of packet loss metric in QoS based network routing
US7499410B2 (en) 2001-12-26 2009-03-03 Cisco Technology, Inc. Fibre channel switch that enables end devices in different fabrics to communicate with one another while retaining their unique fibre channel domain—IDs
US6880025B2 (en) * 2001-12-27 2005-04-12 Koninklijke Philips Electronics N.V. Efficient timeout message management in IEEE 1394 bridged serial bus network
US7099283B2 (en) * 2002-01-25 2006-08-29 Ntt Docomo, Inc. Quality of service aware handoff trigger
US20050120208A1 (en) * 2002-01-25 2005-06-02 Albert Dobson Robert W. Data transmission systems
GB2384663B (en) * 2002-01-25 2004-12-29 Actix Ltd Data transmission systems
US20040042398A1 (en) * 2002-02-28 2004-03-04 Seriqa Networks Method and apparatus for reducing traffic congestion by preventing allocation of the occupied portion of the link capacity and for protecting a switch from congestion by preventing allocation on some of its links
US20050213594A1 (en) * 2002-03-27 2005-09-29 Joachim Klink Method for operation and monitoring of mpls networks
US7616637B1 (en) 2002-04-01 2009-11-10 Cisco Technology, Inc. Label switching in fibre channel networks
US7206288B2 (en) * 2002-06-12 2007-04-17 Cisco Technology, Inc. Methods and apparatus for characterizing a route in fibre channel fabric
GB0220660D0 (en) * 2002-09-05 2002-10-16 Nokia Corp Signal propogation delay routing
US7406082B2 (en) * 2002-09-30 2008-07-29 Lucent Technologies Inc. Sequence number schemes for acceptance/rejection of duplicated packets in a packet-based data network
US7433326B2 (en) 2002-11-27 2008-10-07 Cisco Technology, Inc. Methods and devices for exchanging peer parameters between network devices
US7286482B2 (en) * 2002-11-29 2007-10-23 Alcatel Lucent Decentralized SLS monitoring in a differentiated service environment
US7292537B2 (en) * 2002-11-29 2007-11-06 Alcatel Lucent Measurement architecture to obtain per-hop one-way packet loss and delay in multi-class service networks
JP3799326B2 (en) * 2002-12-02 2006-07-19 Necインフロンティア株式会社 Packet transmission method and packet reception method
US7194661B1 (en) * 2002-12-23 2007-03-20 Intel Corporation Keep alive buffers (KABs)
US7454425B2 (en) * 2003-07-01 2008-11-18 Oracle International Corporation System and method for assembling timing data in a multi-layer server environment
US7391777B2 (en) * 2003-11-03 2008-06-24 Alcatel Lucent Distance-sensitive scheduling of TDM-over-packet traffic in VPLS
US7496661B1 (en) * 2004-03-29 2009-02-24 Packeteer, Inc. Adaptive, application-aware selection of differentiated network services
US7843843B1 (en) 2004-03-29 2010-11-30 Packeteer, Inc. Adaptive, application-aware selection of differntiated network services
US20050281392A1 (en) * 2004-06-18 2005-12-22 Covaro Networks, Inc. System and method for connection performance analysis
JP4390649B2 (en) * 2004-07-14 2009-12-24 富士通株式会社 Network loop detector
US7701852B1 (en) * 2004-08-26 2010-04-20 Sprint Communications Company L.P. Method for analyzing performance of a network through measuring and reporting delay in routing devices
US20070263539A1 (en) * 2004-09-10 2007-11-15 Telefonaktiebolaget Lm Ericsson (Publ) Method For Monitoring Service Performance And Quality Level In A Telecommunications Network
US7443801B2 (en) * 2004-10-28 2008-10-28 Telcordia Technologies, Inc. Remote estimation of round-trip delays in a data network
US7916628B2 (en) * 2004-11-01 2011-03-29 Cisco Technology, Inc. Trunking for fabric ports in fibre channel switches and attached devices
GB2420244A (en) * 2004-11-16 2006-05-17 Agilent Technologies Inc Routing a measurement packet with copy/clone capability dependent upon certain criteria
US7644147B1 (en) * 2005-03-25 2010-01-05 Marvell International Ltd. Remote network device management
US8169891B2 (en) * 2005-03-31 2012-05-01 Agere Systems Inc. Apparatus and method for handling lost cells in a communications system
US20070104188A1 (en) * 2005-11-07 2007-05-10 Zenon Kuc Determining transmission latency in network devices
US7624168B2 (en) * 2006-02-02 2009-11-24 Hostway Corporation Multi-layer system for scalable hosting platform
KR101210341B1 (en) * 2006-02-11 2012-12-10 삼성전자주식회사 Method to accurately and securely measure propagation delay and distance between sending and receiving node in packet network and packet network node for executing the method
US7961635B2 (en) * 2006-05-24 2011-06-14 At&T Intellectual Property I, Lp Network latency analysis packet and method
US8289965B2 (en) 2006-10-19 2012-10-16 Embarq Holdings Company, Llc System and method for establishing a communications session with an end-user based on the state of a network connection
US8194643B2 (en) 2006-10-19 2012-06-05 Embarq Holdings Company, Llc System and method for monitoring the connection of an end-user to a remote network
US8717911B2 (en) 2006-06-30 2014-05-06 Centurylink Intellectual Property Llc System and method for collecting network performance information
US8184549B2 (en) 2006-06-30 2012-05-22 Embarq Holdings Company, LLP System and method for selecting network egress
US8488447B2 (en) 2006-06-30 2013-07-16 Centurylink Intellectual Property Llc System and method for adjusting code speed in a transmission path during call set-up due to reduced transmission performance
US8793361B1 (en) * 2006-06-30 2014-07-29 Blue Coat Systems, Inc. Traffic synchronization across multiple devices in wide area network topologies
US9094257B2 (en) 2006-06-30 2015-07-28 Centurylink Intellectual Property Llc System and method for selecting a content delivery network
US8199653B2 (en) 2006-08-22 2012-06-12 Embarq Holdings Company, Llc System and method for communicating network performance information over a packet network
US7843831B2 (en) 2006-08-22 2010-11-30 Embarq Holdings Company Llc System and method for routing data on a packet network
US8307065B2 (en) 2006-08-22 2012-11-06 Centurylink Intellectual Property Llc System and method for remotely controlling network operators
US8125897B2 (en) 2006-08-22 2012-02-28 Embarq Holdings Company Lp System and method for monitoring and optimizing network performance with user datagram protocol network performance information packets
US8274905B2 (en) 2006-08-22 2012-09-25 Embarq Holdings Company, Llc System and method for displaying a graph representative of network performance over a time period
US8537695B2 (en) 2006-08-22 2013-09-17 Centurylink Intellectual Property Llc System and method for establishing a call being received by a trunk on a packet network
US8144587B2 (en) 2006-08-22 2012-03-27 Embarq Holdings Company, Llc System and method for load balancing network resources using a connection admission control engine
US8576722B2 (en) 2006-08-22 2013-11-05 Centurylink Intellectual Property Llc System and method for modifying connectivity fault management packets
US8743703B2 (en) * 2006-08-22 2014-06-03 Centurylink Intellectual Property Llc System and method for tracking application resource usage
US8107366B2 (en) 2006-08-22 2012-01-31 Embarq Holdings Company, LP System and method for using centralized network performance tables to manage network communications
US8531954B2 (en) * 2006-08-22 2013-09-10 Centurylink Intellectual Property Llc System and method for handling reservation requests with a connection admission control engine
US8223655B2 (en) 2006-08-22 2012-07-17 Embarq Holdings Company, Llc System and method for provisioning resources of a packet network based on collected network performance information
US9479341B2 (en) 2006-08-22 2016-10-25 Centurylink Intellectual Property Llc System and method for initiating diagnostics on a packet network node
US8228791B2 (en) 2006-08-22 2012-07-24 Embarq Holdings Company, Llc System and method for routing communications between packet networks based on intercarrier agreements
US8194555B2 (en) 2006-08-22 2012-06-05 Embarq Holdings Company, Llc System and method for using distributed network performance information tables to manage network communications
US8619600B2 (en) 2006-08-22 2013-12-31 Centurylink Intellectual Property Llc System and method for establishing calls over a call path having best path metrics
US8064391B2 (en) 2006-08-22 2011-11-22 Embarq Holdings Company, Llc System and method for monitoring and optimizing network performance to a wireless device
US8130793B2 (en) 2006-08-22 2012-03-06 Embarq Holdings Company, Llc System and method for enabling reciprocal billing for different types of communications over a packet network
US8750158B2 (en) 2006-08-22 2014-06-10 Centurylink Intellectual Property Llc System and method for differentiated billing
US8238253B2 (en) 2006-08-22 2012-08-07 Embarq Holdings Company, Llc System and method for monitoring interlayer devices and optimizing network performance
US8102770B2 (en) 2006-08-22 2012-01-24 Embarq Holdings Company, LP System and method for monitoring and optimizing network performance with vector performance tables and engines
US8189468B2 (en) 2006-10-25 2012-05-29 Embarq Holdings, Company, LLC System and method for regulating messages between networks
US8015294B2 (en) 2006-08-22 2011-09-06 Embarq Holdings Company, LP Pin-hole firewall for communicating data packets on a packet network
US7684332B2 (en) 2006-08-22 2010-03-23 Embarq Holdings Company, Llc System and method for adjusting the window size of a TCP packet through network elements
US8549405B2 (en) 2006-08-22 2013-10-01 Centurylink Intellectual Property Llc System and method for displaying a graphical representation of a network to identify nodes and node segments on the network that are not operating normally
US8407765B2 (en) 2006-08-22 2013-03-26 Centurylink Intellectual Property Llc System and method for restricting access to network performance information tables
US8144586B2 (en) 2006-08-22 2012-03-27 Embarq Holdings Company, Llc System and method for controlling network bandwidth with a connection admission control engine
US8224255B2 (en) 2006-08-22 2012-07-17 Embarq Holdings Company, Llc System and method for managing radio frequency windows
US8049884B2 (en) * 2006-10-06 2011-11-01 Shimadzu Corporation Spectrophotometer
JP4926648B2 (en) * 2006-10-27 2012-05-09 富士通セミコンダクター株式会社 In-vehicle gateway device
WO2008051123A1 (en) * 2006-10-27 2008-05-02 Telefonaktiebolaget Lm Ericsson (Publ) Method for clock recovery using updated timestamps
US20080159287A1 (en) * 2006-12-29 2008-07-03 Lucent Technologies Inc. EFFICIENT PERFORMANCE MONITORING USING IPv6 CAPABILITIES
KR100773416B1 (en) * 2007-03-21 2007-11-05 (주)소만사 Method and system for controlling network traffic of p2p and instant messenger
US8111692B2 (en) 2007-05-31 2012-02-07 Embarq Holdings Company Llc System and method for modifying network traffic
JP2009153055A (en) * 2007-12-21 2009-07-09 Nippon Hoso Kyokai <Nhk> Relay apparatus, receiver, and transmitter
US7818422B2 (en) * 2007-12-31 2010-10-19 Microsoft Corporation Real-time monitoring of a routing server
US8068425B2 (en) 2008-04-09 2011-11-29 Embarq Holdings Company, Llc System and method for using network performance information to determine improved measures of path states
BRPI0924567A2 (en) * 2009-05-15 2018-10-09 Huawei Tech Co Ltd method, apparatus and system for network performance measurement
EP2436152B1 (en) * 2009-05-29 2013-07-10 Telefonaktiebolaget LM Ericsson (publ) Failure localisation in a mpls-tp network
US20100309908A1 (en) * 2009-06-08 2010-12-09 Hewlett-Packard Development Company, L.P. Method and system for communicating with a network device
US8335163B2 (en) * 2009-10-27 2012-12-18 Microsoft Corporation Quality of service (QOS) based systems, networks, and advisors
US8718482B1 (en) * 2009-11-10 2014-05-06 Calix, Inc. Transparent clock for precision timing distribution
US20110261696A1 (en) 2010-04-22 2011-10-27 International Business Machines Corporation Network data congestion management probe system
US8549201B2 (en) 2010-06-30 2013-10-01 Intel Corporation Interrupt blocker
US9300558B2 (en) * 2010-10-04 2016-03-29 Nec Corporation Delay measurement system and delay measurement method, as well as delay measurement device and delay measurement program
JP5538652B2 (en) * 2010-12-20 2014-07-02 三菱電機株式会社 Network status monitoring method
EP2477357B1 (en) * 2011-01-12 2014-06-25 Alcatel Lucent Traceroute delay diagnostic command
US8909762B2 (en) * 2011-03-17 2014-12-09 Hewlett-Packard Development Company, L.P. Network system management
JP5682391B2 (en) * 2011-03-22 2015-03-11 富士通株式会社 Data transfer apparatus, parallel computer system, and data transfer apparatus control method
US9197691B2 (en) * 2011-10-04 2015-11-24 Google Technology Holdings LLC System and method for latency measurement at each network element participating as an RTP relay in a telecommunication network
US9252903B2 (en) * 2011-10-21 2016-02-02 Cortina Systems, Inc. System and method for accounting for time that a packet spends in transit through a transparent clock
JP5789548B2 (en) * 2012-03-16 2015-10-07 株式会社日立製作所 Communication device
CN102685014B (en) * 2012-05-29 2014-06-25 华为技术有限公司 Method for measuring performance index of service flow and sending terminal equipment
US8848563B2 (en) 2012-06-12 2014-09-30 Calix, Inc. Systems and methods for measuring frame loss in multipoint networks
US8989032B2 (en) 2012-06-12 2015-03-24 Calix, Inc. Systems and methods for measuring frame loss in multipoint networks
US9407733B1 (en) * 2012-11-09 2016-08-02 Marvell Israel (M.I.S.L) Ltd. Time correction using extension fields
US9509585B1 (en) * 2013-09-03 2016-11-29 Xpliant, Inc. Apparatus and method for time stamping packets across several nodes in a network
WO2015101952A1 (en) * 2014-01-02 2015-07-09 Marvell World Trade Ltd Accurate measurement of distributed counters
US9300565B2 (en) * 2014-04-17 2016-03-29 Accedian Networks Inc. System and method for out-of-line real-time in-service performance measurement
US10511498B1 (en) 2015-02-25 2019-12-17 Infoblox Inc. Monitoring and analysis of interactions between network endpoints
US11102273B2 (en) * 2015-05-13 2021-08-24 Cisco Technology, Inc. Uplink performance management
JP6572136B2 (en) * 2016-01-13 2019-09-04 エヌ・ティ・ティ・コミュニケーションズ株式会社 COMMUNICATION SYSTEM, COMMUNICATION DEVICE, SECOND DEVICE, COMMUNICATION METHOD, AND COMPUTER PROGRAM
US10567290B2 (en) 2016-02-08 2020-02-18 Telefonaktiebolaget Lm Ericsson (Publ) Method and switch for managing traffic in transport network
US10651974B2 (en) 2017-02-28 2020-05-12 Marvell Asia Pte, Ltd. Method and apparatus for updating error detection information in packets

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5251204A (en) * 1990-09-19 1993-10-05 Fujitsu Limited Transmission test system in a broadband ISDN
US5313453A (en) * 1991-03-20 1994-05-17 Fujitsu Limited Apparatus for testing ATM channels
US5450394A (en) * 1994-03-10 1995-09-12 Northern Telecom Limited Delay monitoring of telecommunication networks
US5457700A (en) * 1992-08-17 1995-10-10 At&T Ipm Corp. Asynchronous transfer mod (ATM) transmission test cell generator
US5504754A (en) * 1993-04-08 1996-04-02 Alcatel Str A.G. Method and arrangement for the load testing of a switch for cell structured digital data

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0528075A1 (en) * 1991-08-19 1993-02-24 ALCATEL BELL Naamloze Vennootschap Performance measurement device for a telecommunication path and method used therein
GB2261799B (en) * 1991-11-23 1995-04-19 Dowty Communications Ltd Packet transmission system
GB2267200B (en) * 1992-05-19 1995-10-25 Dowty Communications Ltd Packet transmission system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5251204A (en) * 1990-09-19 1993-10-05 Fujitsu Limited Transmission test system in a broadband ISDN
US5313453A (en) * 1991-03-20 1994-05-17 Fujitsu Limited Apparatus for testing ATM channels
US5457700A (en) * 1992-08-17 1995-10-10 At&T Ipm Corp. Asynchronous transfer mod (ATM) transmission test cell generator
US5504754A (en) * 1993-04-08 1996-04-02 Alcatel Str A.G. Method and arrangement for the load testing of a switch for cell structured digital data
US5450394A (en) * 1994-03-10 1995-09-12 Northern Telecom Limited Delay monitoring of telecommunication networks

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
EUROPEAN NETWORK LABORATORIES, issued March 1995, ROBERT MANDEVILLE, "The ATM Stress Test: Which Switches Survived", pages 1-15. *
See also references of EP0892958A4 *

Cited By (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2785117A1 (en) * 1998-10-21 2000-04-28 Celogic Client server communications network link validation/qualification/fault detection having fault detection discriminating anomalies and analysis finding anomaly/fault origin/nature.
FR2785115A1 (en) * 1998-10-21 2000-04-28 Celogic Client server communications network link validation/qualification/fault detection having fault detection discriminating anomalies and analysis finding anomaly/fault origin/nature.
WO2000024223A1 (en) * 1998-10-21 2000-04-27 Celogic Method for data qualification and for assisting in fault localization on a digital telecommunication network
GB2381696B (en) * 2001-10-24 2004-10-13 Zarlink Semiconductor Vn Inc Methods of performance estimation in provisioning delay intolerant data services
US8612625B2 (en) 2009-01-22 2013-12-17 Candit-Media Characterizing data flow in a network based on a footprint measurement and taking action to avoid packet loss including buffer scaling or traffic shaping by adapting the footprint measurement
EP2211508A1 (en) * 2009-01-22 2010-07-28 IBBT vzw Method and device for characterising a data flow in a network for transferring media data
WO2010084151A1 (en) 2009-01-22 2010-07-29 Ibbt Vzw Method and device for characterising a data flow in a network for transferring data
US8750356B2 (en) 2009-11-27 2014-06-10 Huawei Technologies Co., Ltd. Method, apparatus, and system for measuring asymmetric delay of communication path
EP2458787A4 (en) * 2009-11-27 2012-05-30 Huawei Tech Co Ltd Method, device and system for measuring asymmetric time delay of communication paths
EP2458787A1 (en) * 2009-11-27 2012-05-30 Huawei Technologies Co., Ltd. Method, device and system for measuring asymmetric time delay of communication paths
GB2476116A (en) * 2009-12-14 2011-06-15 Dublin Inst Of Technology Real-time VoIP transmission quality predictor and quality-driven de jitter buffer
EP2381622A1 (en) * 2010-04-23 2011-10-26 Alcatel Lucent Update of a cumulative residence time of a packet in a packet-switched communication network
CN102859941B (en) * 2010-04-23 2016-07-06 阿尔卡特朗讯 Renewal to the accumulative residence time being grouped in packet exchange communication network
JP2013526167A (en) * 2010-04-23 2013-06-20 アルカテル−ルーセント Updating the accumulated packet residence time in packet-switched communication networks
US9203725B2 (en) 2010-04-23 2015-12-01 Alcatel Lucent Update of a cumulative residence time of a packet in a packet-switched communication network
WO2011131556A1 (en) * 2010-04-23 2011-10-27 Alcatel Lucent Update of a cumulative residence time of a packet in a packet-switched communication network
CN102859941A (en) * 2010-04-23 2013-01-02 阿尔卡特朗讯 Update of a cumulative residence time of a packet in a packet-switched communication network
EP2562963A1 (en) * 2011-08-25 2013-02-27 Honeywell International Inc. Embedded end-to-end delay information for data networks
US8665884B2 (en) 2011-08-25 2014-03-04 Honeywell International Inc. Embedded end-to-end delay information for data networks
WO2013098541A1 (en) 2011-12-29 2013-07-04 British Telecommunications Plc Distributed system management
US11218534B2 (en) 2011-12-29 2022-01-04 British Telecommunications Public Limited Company Distributed system management
US11431785B2 (en) 2011-12-29 2022-08-30 British Telecommunications Public Limited Company Distributed system management
WO2013098543A1 (en) 2011-12-29 2013-07-04 British Telecommunications Public Limited Company Distributed system management
WO2013098542A1 (en) 2011-12-29 2013-07-04 British Telecommunications Public Limited Company Distributed system management
EP2611069A1 (en) * 2011-12-29 2013-07-03 British Telecommunications Public Limited Company Distributed system management
US8964755B2 (en) 2011-12-30 2015-02-24 British Telecommunications Public Limited Company Obtaining information from data items
EP2648099A1 (en) * 2012-02-24 2013-10-09 British Telecommunications public limited company Distributed system management
EP2648365A1 (en) * 2012-02-24 2013-10-09 British Telecommunications public limited company Distributed system management
EP2637370A1 (en) * 2012-02-24 2013-09-11 British Telecommunications public limited company Distributed system management

Also Published As

Publication number Publication date
JP3894232B2 (en) 2007-03-14
EP0892958A4 (en) 2000-04-05
CA2250278C (en) 2004-08-24
JP2000507779A (en) 2000-06-20
US5793976A (en) 1998-08-11
CA2250278A1 (en) 1997-10-09
EP0892958A1 (en) 1999-01-27

Similar Documents

Publication Publication Date Title
US5793976A (en) Method and apparatus for performance monitoring in electronic communications networks
US5761191A (en) Statistics collection for ATM networks
US5699346A (en) Measuring burst rate and burst size in ATM network virtual connections
US5812528A (en) Measuring round trip time in ATM network virtual connections
US5764626A (en) Rate-matched cell identification and modification, replacement, or insertion for test and measurement of ATM network virtual connections
US5373504A (en) Apparatus and a method for setting a communication path
US5878032A (en) Delay monitoring of telecommunication networks
US5544169A (en) Apparatus and a method for supervising and controlling ATM traffic
US6654923B1 (en) ATM group protection switching method and apparatus
US6198726B1 (en) Plural-line terminating apparatus and OAM processing method thereof
US6141326A (en) Exchange having function for detecting fault in internal unit
Sykas et al. Overview of ATM networks: functions and procedures
Chen et al. Monitoring and control of ATM networks using special cells
WO2000056116A1 (en) Method and apparatus for performing packet based policing
US6791943B1 (en) Event-based technique for determining instantaneous cell bandwidth in a digital communications network
JP2000059370A (en) Device and method for controlling traffic
US6898177B1 (en) ATM protection switching method and apparatus
EP0861534B1 (en) Instrument for test and measurement of atm network virtual connections
US6665302B2 (en) Method and system for handling a loop back connection using a priority unspecified bit rate in ADSL interface
EP1093263A2 (en) Method and devices for cell loss detection in ATM telecommunication devices
De Prycker et al. An ATM switching architecture with intrinsic multicast capabilities for the Belgian broadband experiment
Saito Bandwidth management for AAL2 traffic
Kelley Comprehensive in-service measurement of QoS
CA2459468C (en) Instrument for test and measurement of atm network virtual connections
Toyoshima et al. Flexible surveillance capabilities for ATM-based transmission systems

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): CA JP

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
ENP Entry into the national phase

Ref document number: 2250278

Country of ref document: CA

Ref country code: CA

Ref document number: 2250278

Kind code of ref document: A

Format of ref document f/p: F

WWE Wipo information: entry into national phase

Ref document number: 1997917052

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 1997917052

Country of ref document: EP