CN1625170B - Method for realizing large number of user on-line simultaneously of universal group radio operation - Google Patents
Method for realizing large number of user on-line simultaneously of universal group radio operation Download PDFInfo
- Publication number
- CN1625170B CN1625170B CN 200310120070 CN200310120070A CN1625170B CN 1625170 B CN1625170 B CN 1625170B CN 200310120070 CN200310120070 CN 200310120070 CN 200310120070 A CN200310120070 A CN 200310120070A CN 1625170 B CN1625170 B CN 1625170B
- Authority
- CN
- China
- Prior art keywords
- memory
- group
- data
- service
- slpi
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Images
Abstract
This invention discloses a method realizing a large number of users online simultaneously of the general grouping radio operation. This method is: monitor the state of the example automatic machine ofthe operation logic processing program; when the time of the leisure stated of the automatic machine surpasses the pre-set time, the data control and managing entity stores the internal data of the automatic machine and releases the EMS memory ; when receiving the information of applying fare registering report or invent report, and if finding the internal data of the corresponding example of theoperation logic processing program being stored, pick up the internal data from the relating data control and managing entity to the memory and process the next operation logic.
Description
Technical field
The present invention relates to intelligent network GPRS (General Packet Radio Service is called for short GPRS), relate in particular to the simultaneously online implementation method of GPRS a large number of users.
Background technology
GPRS intelligent network principle is consulted shown in Figure 1, Serving GPRS Support Node (Serving GPRS SupportNode, be called for short SGSN) be in global system for mobile communications (the being called for short GSM) network in order to support the GPRS function newly-increased equipment, gsm wireless Network controlled function (GSM Service Control Function, be called for short gsmSCF) by service control point (Service Control Point, be called for short SCP) provide, GPRS wireless network traffic function of exchange (GPRS Service Switching Function, be called for short GprsSSF) can be embedded among the SGSN or provide by SSP independently, pass through gprsSSF, signaling between the gsmSCF cooperates, and finishes subscription authentication, call supervision, functions such as charging.Details see also 3GPP TS 23.078.
The model of gsmSCF as shown in Figure 2, wherein " Service Logic processing Program Instance " is the key modules of finishing subscription authentication, call supervision, charging, usually corresponding one " Service Logic processing Program Instance " of intelligent call.
If processing method according to traditional IN service, after SCP receives an intelligent call, create a Service Logic processing Program Instance (Service Logic Processing Programme Instance, hereinafter to be referred as SLPI), this SLPI terminate-and-stay-resident in call handling process, discharge this SLPI behind the end of calling, so for the simultaneously online situation of a large amount of (as 1,000,000) gprs service users, 1,000,000 SLPI will consume the internal memory (a common SLPI needs the 4K internal memory) of 10000000 * 4K ≈ 4G, add the internal memory that other module consumes, SCP need dispose the above physical memory of 5G.And the SCP equipment of widely applying at present; major part all is to adopt the standard configuration of minicomputers such as HP/SUN/IBM; all do not dispose the above internal memory of 5G; the problem of bringing like this is exactly; on the SCP of existing network equipment, open gprs service; need shutdown to add internal memory, influence service quality, also can make operator increase equipment investment.
In order to address the above problem, the method that a kind of intelligent network gprs service supports that a large number of users is simultaneously online is proposed in the industry, the key technology of this method is by the technology of file storage sluggish SLPI automaton in the internal memory to be exchanged on the hard disk, vacate a large amount of unallocated internal memories and use for new calling, thereby realize that a large number of users is simultaneously online.
Consult Fig. 2, its principle is: monitored by the state of the management entity of the service control function on the Service Control Function-control (SCME-Control) to Service Logic processing Program Instance (SLPI) automaton of processing gprs service.Surpass set point when the free time of any Service Logic processing Program Instance (SLPI) automaton, service control function management entity-control (SCME-Control) is cached to its memory information on the disk and discharges internal memory.When Service Control Function is received the GPRS Apply Charging Report of Serving GPRS Support Node (SGSN) or GPRS event report message, service control function management entity-control (SCME-Control) has been cached to disk if find corresponding service logical process program example (SLPI), then from disk, read corresponding information in internal memory, and carry out follow-up business logic processing.
Said method has smoothly solved because the online memory consumption problem that produces of a large number of users while by the technology of file cache.But what the file cache technology adopted is the mutual technology of same External memory equipment (as external equipments such as hard disk, tapes, being called for short " external memory " here), and external memory has the characteristics of big capacity, low-response.Though external memory big capacity characteristics have solved the mutual simultaneously and problem of the memory consumption that produces of gprs service a large number of users, the slow shortcoming of the response speed of external memory has also influenced the performance of whole SCP system.When the gprs service a large number of users was simultaneously online, the operation meeting of disk sharply increased in the SCP system, and is sharply slack-off to the response meeting of outside other system/entity (as SGSN) like this.So said method, provide good solution to the relatively large user of gprs service is simultaneously online, but when reaching 1,000,000 millions along with the increase of number of users, the defective of this technology on performance also can be by rapid amplification, thereby has limited its application.
Summary of the invention
The object of the present invention is to provide the simultaneously online implementation method of a kind of GPRS a large number of users, when adopting the exterior storage technology to realize that a large number of users is simultaneously online, have the problem that influences systematic function to solve prior art.
For addressing the above problem, the invention provides following technical scheme:
The implementation method that a kind of GPRS a large number of users is simultaneously online is provided for realizing the Data Control management entity (DCME) of mass data buffer memory and abstraction function on the Service Control Function of described intelligent network, the method comprising the steps of:
State to Service Logic processing Program Instance (SLPI) automaton on SCF monitors;
When the time that any Service Logic processing Program Instance (SLPI) automaton is in idle condition surpasses predetermined value, by described Data Control management entity with the memory information buffer memory of this Service Logic processing Program Instance and the internal memory of release busy;
When SCF receives Apply Charging Report or event report message, be buffered if find the memory information of corresponding service logical process program example (SLPI), then extract this memory information to internal memory and carry out follow-up business logic processing from described Data Control management entity.
According to said method:
Described Data Control management entity adopts the memory block management group to come the described memory information of buffer memory; Perhaps adopt blocks of files management group that described memory information is cached to external memory; Perhaps adopt single file management group that described memory information is cached to external memory.
Described Data Control management entity adopts memory block management group, blocks of files management group and single file management group or three's combination to come the described memory information of buffer memory.
Described memory block group is as data fast cache district, by alternately data being swapped out to external memory or changing to the memory block group from external memory with the blocks of files group.
Memory block group can only corresponding blocks of files group, and a blocks of files group can corresponding a plurality of memory block groups.
Described Data Control management entity (DCME) compresses laggard row cache with described memory information, and when extracting this memory information it is decompressed.
Described Data Control management entity adopts fixing compression algorithm; Perhaps according to described memory information feature Dynamic Selection compression algorithm.
The present invention has following beneficial effect:
1, swaps out the memory information of buffer memory Service Logic processing Program Instance by changing to of DCME, thereby avoided frequent I/O operation; Simultaneously, by DCME can optimized using system hardware resource, solve Large Volume Data memory consumption and the excessive problem of performance consumption down, realization " 1,000,000 grades of other a large amount of GPRS users are online ".
2, can keep compatible consistency with original technology by DCME.
3, can be by DCME according to the complexity situation of system, select the technical scheme of a kind of cache management pattern, and on interface, keep consistency and do not change as " a large number of users is online ", can satisfy the demand of software follow-up developments.
Description of drawings
Fig. 1 is a GPRS intelligent network schematic diagram;
Fig. 2 is that present technology realizes the simultaneously online schematic diagram of intelligent network gprs service support a large number of users;
Fig. 3 is the model schematic diagram of intelligent network business controlled function among the present invention;
Fig. 4 is the signaling process figure of a packet data call of the present invention;
Fig. 5 is the logic schematic diagram of DCME of the present invention.
Embodiment
Consult shown in Figure 3ly, (Data-Control-Manage-Entity DCME) realizes the function of the preservation and the extraction of mass data to increase functional entity-Data Control management entity in SCF.This functional entity can be an enhancement function on the SCME-Control entity, realizes the function of Data Control management; Also can be an independence and concrete module entity, realize the function of Data Control management.
The calling of this class Packet Service of GPRS has the characteristics of " the mutual time is short, and the stand-by period is long ".Though it is very long usually to call out line duration, but in the online process of user, gsmSCF with the Signalling exchange between the gprsSSF seldom, and the mutual duration is all very short, the SLPI of overwhelming majority time gsmSCF (be used for calling of unique expression) is in idle condition.
Consult Fig. 3, in the SCF running, service control function management entity-control (SCME-Control) gives a timer monitoring to each mutual Service Logic processing Program Instance (SLPI), does not think that then this SLPI is in " not mutual " state and then can enters " free time " state alternately in the time range in certain appointment.In a single day this SLPI automaton enters after the idle condition, and SCF preserves the memory information of corresponding SLPI by the functional interface of DCME.When SCF receives GPRS Apply Charging Report (ApplyChargingReportGPRS), the GPRS event report message such as (EventReportGPRS) of SGSN, if finding corresponding SLPI is monitored by DCME, then extract the data of this SLPI, and recover SLPI by the interface of DCME; Carry out follow-up business logic processing then, idle Ti is after the time again as SLPI, and SCF preserves the memory information of this SLPI once more by the functional interface of DCME.
Consult Fig. 4, the figure shows the signaling process that a typical pre-payment GPRS user initiates a packet data call.After SCP receives that GPRS starts DP (InitialDPGPRS) message, trigger the pre-payment gprs service, after business is finished subscription authentication, require SGSN configuration DP incident by request report GPRS incident (RequestReportGPRSEvent) message, time and the flow that can surf the Net according to the user balance budget then, by two GPRS Apply Charging (ApplyChargingGPRS) message, but be handed down to SGSN but carry surf time and use traffic respectively, require SGSN to monitor, issue ContinueGPRS message subsequently and allow SGSN carry out subsequent treatment, make the user be linked into Internet.Issue after the ContinueGPRS, service logic on the SCP is suspended, and the SLPI automaton has just entered idle condition, and idle Ti is after the time, and SCME-Control preserves the memory information of this SLPI by the functional interface of DCME.After the user is off line, SGSN reports two GPRS Apply Charging Report (ApplyChargingReportGPRS) message, carry the time of user's online and the flow that has used respectively, when SCF receives ApplyChargingReportGPRS, find that corresponding SLPI has been saved in DCME, then recover SLPI again, carry out the follow-up logical process of deducting fees then, issue GPRS release (ReleaseGPRS) message then and discharge packet data call to SGSN by extracting data in the functional interface of DCME.
Consult shown in Figure 5ly, the logical construction of DCME is made up of 4 parts: interface, compressed and decompressed module, memory management and external memory are managed.
Of paramount importance is the high-speed cache that is combined into a little memory block by compressed and decompressed module and memory management group, realizes the storage of mass data, can solve the problem of performance and internal memory.
Interface: the interface that " preservation data " and " extraction data " are provided.DCME externally provides unified interface.
Compressed and decompressed module: multiple compression, decompression algorithm are provided, are used to remove redundant data.Can also can be according to " data characteristics " Dynamic Selection compression algorithm in this part with the clear and definite compression algorithm of hard coded mode for compression algorithm; The strategy of selecting is " maximum compression rate ", " the fastest compression speed ", " comprehensive optimum " etc., can adopt to obtain in the system configuration information and do clear and definite decision-making.
Memory management: realize the management of one group of memory block according to the memory block size, be used for the data after the buffer memory compression.In the system different memory block groups can be arranged, as 1024byte, 2048byte, 4096byte ....What of memory block management group, the size of the memory block of every group of management can be obtained from the configuration information of system.The memory block group is as the fast cache district of data.
File management: file management is similar with the realization of memory management, and the blocks of files size also can be obtained from the configuration information of system, and multiple composite type is also arranged.
There are a kind of corresponding relation in blocks of files and memory block: memory block group can only corresponding blocks of files group;
A blocks of files group can corresponding a plurality of memory block groups.
Single file management group: the data after compressed and decompressed can realize management by memory management group and file management group, also can be directly be saved in the external memory medium with the form (each data block is saved in an independent file) of single file.
Preserve the process of data: " preservation data " interface by DCME receives a blocks of data; According to compression algorithm of policy selection of current system, remove section data redundancy (this operation can make the size of data shrink at double) then to the algorithm selection; Then in " memory management " module according to selection strategy (normally selecting) according to the size of data after the compression, select a memory block management group to preserve this data; The memory block management group is mutual with blocks of files management group by certain strategy (ratio or first in first out (FIFO) algorithm that take as freed data blocks), and data are swapped out in the external memory.
Extract the process of data: by " extraction data " interface of DCME, obtain the information of data of having preserved, preserve the data position obtains compression from " the memory block management group " or " blocks of files management group " of preserving this data block after according to the data in the information then; Then according to using corresponding decompression algorithm restore data.Recovered data is exactly the initial data that will extract.
The Compress softwares compression algorithm is an optional process, promptly can ignore this process; Thereby can effectively remove the redundant utilance that improves internal memory of section data by compression algorithm, the Compress softwares compression algorithm can be selected the algorithm of a kind of performance and space optimum according to data characteristics simultaneously.
Memory management is the memory block management group of the different big or small types of a group.Memory block number in every kind of memory block management group can be the number of static configuration, also can dynamically determine the memory block number according to the service feature of current system.
Single file management group essence is a degradation model of " memory management group ", i.e. processing under the extreme case of memory management group.The process of this processing also is compatible an assurance to original invention.
Application by " memory management group " can realize " the gprs service user is online in a large number " substantially, can effectively solve the simultaneously online at hundred general-purpose families of performance resource and memory source and the contradiction that produces.
Flexible framework by DCME can be selected part frame according to the application difficulty of system, and promptly DCME develops and is degenerated to corresponding comparatively simple frame model:
Interface+compression/de-compression module+memory management group+file management group+single file management group;
Interface+compression/de-compression module+memory management group+file management group;
Interface+compression/de-compression module+memory management group+single file management group;
Interface+compression/de-compression module+file management group+single file management group;
Interface+compression/de-compression module+memory management group;
Interface+compression/de-compression module+single file management group;
Interface+compression/de-compression module+file management group;
Interface+file management group;
Interface+memory management group;
Interface+single file management group.
In the present invention, also can monitor the idle condition of Service Logic processing Program Instance (SLPI) by other modes.
Claims (8)
1. the simultaneously online implementation method of a GPRS a large number of users, it is characterized in that, be provided for realizing the Data Control management entity (DCME) of mass data buffer memory and abstraction function on the Service Control Function of intelligent network, the method comprising the steps of:
State to Service Logic processing Program Instance (SLPI) automaton on SCF monitors;
When the time that any Service Logic processing Program Instance (SLPI) automaton is in idle condition surpasses predetermined value, by described Data Control management entity with the internal storage data buffer memory of this Service Logic processing Program Instance and the internal memory of release busy;
When SCF receives Apply Charging Report or event report message, be buffered if find the memory information of corresponding service logical process program example (SLPI), then extract this memory information to internal memory and carry out follow-up business logic processing from described Data Control management entity.
2. the method for claim 1 is characterized in that, described Data Control management entity adopts the memory block management group to come the described memory information of buffer memory; Perhaps adopt blocks of files management group that described memory information is cached to external memory; Perhaps adopt single file management group that described memory information is cached to external memory.
3. the method for claim 1 is characterized in that, described Data Control management entity adopts memory block management group, blocks of files management group and single file management group or three's combination to come the described memory information of buffer memory.
4. the method for claim 1 is characterized in that, adopts the memory block group as data fast cache district, by alternately data being swapped out to external memory or changing to the memory block group from external memory with the blocks of files group.
5. method as claimed in claim 3 is characterized in that, memory block group can only corresponding blocks of files group, and a blocks of files group can corresponding a plurality of memory block groups.
6. as the arbitrary described method of claim 1 to 5, it is characterized in that described Data Control management entity (DCME) compresses laggard row cache with described memory information, and when extracting this memory information, it is decompressed.
7. method as claimed in claim 6 is characterized in that, described Data Control management entity adopts fixing compression algorithm; Perhaps according to described memory information feature Dynamic Selection compression algorithm.
8. the method for claim 1 is characterized in that, is monitored by the state of the management entity of the service control function on the SCF-control (SCME-Control) to Service Logic processing Program Instance (SLPI) automaton.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200310120070 CN1625170B (en) | 2003-12-03 | 2003-12-03 | Method for realizing large number of user on-line simultaneously of universal group radio operation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200310120070 CN1625170B (en) | 2003-12-03 | 2003-12-03 | Method for realizing large number of user on-line simultaneously of universal group radio operation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1625170A CN1625170A (en) | 2005-06-08 |
| CN1625170B true CN1625170B (en) | 2010-04-28 |
Family
ID=34761491
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200310120070 Expired - Fee Related CN1625170B (en) | 2003-12-03 | 2003-12-03 | Method for realizing large number of user on-line simultaneously of universal group radio operation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1625170B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100442782C (en) * | 2005-11-04 | 2008-12-10 | 华为技术有限公司 | Method for improving packet service on-line user number |
| CN107273210A (en) * | 2017-06-13 | 2017-10-20 | 上海青橙实业有限公司 | The management method and device of Jack Server examples |
| CN109460291A (en) * | 2018-09-26 | 2019-03-12 | 平安科技（深圳）有限公司 | Electronic device, the method for controller perturbation server cluster and storage medium |
| TWI745697B (en) | 2019-05-24 | 2021-11-11 | 創鑫智慧股份有限公司 | Computing system and compressing method thereof for neural network parameters |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1225216A (en) * | 1996-07-19 | 1999-08-04 | 艾利森电话股份有限公司 | Dynamic load limiting |
| US6574201B1 (en) * | 1998-07-06 | 2003-06-03 | Siemens Aktiengesellschaft | Method and mobile radio telephone network for handling a packet data service |
-
2003
- 2003-12-03 CN CN 200310120070 patent/CN1625170B/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1225216A (en) * | 1996-07-19 | 1999-08-04 | 艾利森电话股份有限公司 | Dynamic load limiting |
| US6574201B1 (en) * | 1998-07-06 | 2003-06-03 | Siemens Aktiengesellschaft | Method and mobile radio telephone network for handling a packet data service |
Non-Patent Citations (6)
| Title |
|---|
| . |
| .2002,43-46 |
| .大连海事大学硕士学位论文 |
| .智能网业务控制点负载平衡的研究 |
| 张炯 |
| 张炯.智能网业务控制点负载平衡的研究.大连海事大学硕士学位论文.2002,43-46. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1625170A (en) | 2005-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101588602B (en) | Method for reducing power consumption of WAPI mobile terminal and a WAPI mobile terminal | |
| CN104239132B (en) | A kind of wake up the method for alignment, device and terminal up | |
| CN101360075B (en) | Grouped gateway and method for saving power consumption | |
| CN103324549A (en) | Method for realizing sleep standby and communication terminal | |
| CN100454829C (en) | Log processing system and method | |
| CN1625170B (en) | Method for realizing large number of user on-line simultaneously of universal group radio operation | |
| CN105162622A (en) | Storage method and system | |
| CN101951623A (en) | User behavior statistical method and device based on user events | |
| CN101351045B (en) | Method, equipment for marking terminal PRI and equipment for identifying terminal PRI | |
| CN106557434A (en) | A kind of interface caching method and system | |
| CN100518359C (en) | System and method for processing the signaling message report | |
| CN100493000C (en) | Method and device for realizing multiple logic path counting | |
| CN111796935A (en) | Consumption instance distribution method and system for calling log information | |
| CN101102176A (en) | A data backup method | |
| CN110297785A (en) | A kind of finance data flow control apparatus and flow control method based on FPGA | |
| CN1332534C (en) | Method of supporting node monitoring mobile user state by service universal packet wireless service | |
| CN110069457A (en) | One kind is based on scratch file delet method and system in distributed file storage system | |
| CN102147735A (en) | Interface single board and business logic loading method | |
| CN102056108B (en) | Short message cache method and device and telecommunication smart card | |
| CN108241487A (en) | The system architecture of mobile equipment performance can be improved | |
| CN111654886A (en) | Method and device for limiting user bandwidth | |
| CN103731313B (en) | Counter and its implementation based on DDR SDRAM | |
| CN111737297A (en) | Method and device for processing link aggregation call information | |
| CN102573021B (en) | The method that wireless sensor network message is processed and gateway | |
| CN100442708C (en) | Method and system for active/standby backup |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100428 Termination date: 20121203 |