CN100548072C - The system that is used for distributed call processing and online upgrading - Google Patents

The system that is used for distributed call processing and online upgrading Download PDF

Info

Publication number
CN100548072C
CN100548072C CN 02159599 CN02159599A CN100548072C CN 100548072 C CN100548072 C CN 100548072C CN 02159599 CN02159599 CN 02159599 CN 02159599 A CN02159599 A CN 02159599A CN 100548072 C CN100548072 C CN 100548072C
Authority
CN
China
Prior art keywords
call
server application
group
backup
application program
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
Application number
CN 02159599
Other languages
Chinese (zh)
Other versions
CN1434648A (en
Inventor
戴维·A·柯林斯
罗伊·C·戈斯韦尔
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Samsung Electronics Co Ltd
Original Assignee
Samsung Electronics Co Ltd
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
Priority claimed from US10/038,879 external-priority patent/US6917819B2/en
Priority claimed from US10/039,186 external-priority patent/US6862453B2/en
Priority claimed from US10/038,878 external-priority patent/US6947752B2/en
Priority claimed from US10/038,872 external-priority patent/US7366521B2/en
Priority claimed from US10/100,494 external-priority patent/US20030176163A1/en
Priority claimed from US10/174,338 external-priority patent/US7308491B2/en
Application filed by Samsung Electronics Co Ltd filed Critical Samsung Electronics Co Ltd
Publication of CN1434648A publication Critical patent/CN1434648A/en
Application granted granted Critical
Publication of CN100548072C publication Critical patent/CN100548072C/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Images

Landscapes

  • Mobile Radio Communication Systems (AREA)
  • Telephonic Communication Services (AREA)

Abstract

A kind of switch, caller device and the calling between the called device that can handle on many trunk lines relevant with described switch connect, described switch comprises: Main Processor Unit, can carry out the call treatment client applications, wherein, each described call treatment client applications is relevant with one of described calling connection; With N call applications node, can carry out the call processing server application program, wherein, the first call processing server application program is carried out on the first call applications node of described N call applications node, relevant with the similar second call processing server application program of on the second call applications node of described N call applications node, carrying out, described second call applications node and the described first call applications node are separated, thereby the described first and second call processing server application programs form the first load Sharing group server application.

Description

The system that is used for distributed call processing and online upgrading
Technical field
The present invention relates generally to communication system, the particularly a kind of distributed call processing of working load shared group and system and method for online upgrading of being used for.
Background technology
There is the client who surpasses 300,000,000 cell phones and other wireless device in the whole world at present.These wireless devices of significant proportion are used as " data pipe " (that is, speech business is not a basic function).In the U.S., the cellular service is provided by cellular service provider, regional Bell company and national distance operator.Keen competition forces the price of cellular service to drop to the stage that most people can bear.
Current this generation cell phone is mainly used in by wireless network and carries out voice conversation between user's set (or wireless device) and the opposing party.A spot of wireless device is a data set, such as the PDA(Personal Digital Assistant) that is equipped with honeycomb/radio modem.Because the bandwidth of current this generation wireless device is limited to tens kilobits per seconds (Kbps) usually, the application of current this generation wireless device is limited relatively.Yet, wish to change this point in for honeycomb/wireless technology next (or 3rd), next (or 3rd) for honeycomb/wireless technology be called as sometimes " 3G " wireless/honeycomb, wherein, each wireless device can use bigger bandwidth (that is 125Kbps or higher).The internet, applications that higher data transfer rate will be used in wireless device is more general.For example, the 3G cell phone PC of 3G cellular modem (or have) can be used for the view Internet website, sends and receive figure, carries out streaming audio or Video Applications etc.The handled wireless traffic of more a high proportion of 3G cellular system will be Internet protocol (IP) business, and the shared ratio of traditional voice service will be lower.
The real-time stream that transmits content of multimedia through Internet protocol (IP) network becomes more general application in recent years.As mentioned above, the 3G wireless network will be provided for the stream data (video and audio frequency) of application in real time to wireless device." " data flow in real time that interactive widely and non-interaction multimedia internet, applications such as news on demand, TV live telecast, video conferencing, outside broadcasting (such as Broadcast.com) etc., will provide to wireless device.Not resembling first warp " non real-time " obtains, and then watch or " download " video file of playing such, (or streaming) data application requirements data source coding streaming data-signal and send to receiver through network in real time, receiver must real-time decoding and play signal (video or audio frequency).
Traditional travelling carriage comprises the larger switch structure by Main Processor Unit (MPU) control usually, and Main Processor Unit comprises mass data processor and relevant memory, and it often occurs with the asic chip form.Each MPU processor comprises the call treatment client applications, and it is used to control the control signal stream of individual call.Each call treatment client applications successively with call processing server interapplication communications, a large amount of control signals streams of calling out of call processing server application program control.
Like this, when particular event (for example, call setup, enable third-party call, call out to disconnect etc.) took place during call, the control signal relevant with this incident was relayed to the call treatment client applications of mobile switching centre (MSC) from travelling carriage.Then, this call treatment client applications is relayed to the call processing server application program with this control signal, the call treatment service that the actual execution of call processing server application program control signal is asked.
Regrettably, in hicap, around the call processing server application program bottleneck may appear.Each call treatment client applications must be communicated by letter with the specific portion of the server hardware of carrying out the call processing server application program.Because the beginning of call and the randomness that stops, so in big system, some servers may and bottleneck occur near the limit of capacity, and other server still has enough bandwidth.In addition, the out of order system of the specific portion of server hardware causes losing of all-calls processing, and this call treatment is to be handled by the call processing server application program that failed server is carried out.
For example, in traditional call handler application program, distributed trunk line mutually relevant with the call state data.When initiating new calling, distribute idle trunk with the trunkfree listserv, then that it is relevant with the call state data.When end of calling, discharge the trunk line that is distributed.Conventional method is used the individual server management free time and is distributed trunk line and related data, this individual server to safeguard the distribution and the release of trunk line, and supports reading and writing of related data.These methods have all been represented a failure part.,, use backup server to take over the work of master server so that make when the master server fault by using backup server to improve individual server.Yet this method is a bottleneck on systematic function, and its autgmentability is not strong.Along with increasing call business by switch management, thereby the performance of individual server has just reduced.
In addition, when new calling begins (technical term is " initiation "), need in the whole life cycle of calling out, should call out by unique identification.Unique identification is called as " call identification ", is the numeral unique to this calling.If in the life cycle of calling out, take place any incident (such as triggering reply, various overtime, on-hook etc.) time, this incident is passed to the concrete call handler application program of calling out of management.The call identification value is used for locating events, and it is transmitted to correct call handler application program.Traditional system distributes the call identification value with individual server.The server of prior art distributes Any Digit when being requested, and it is labeled as uses.When end of calling, the call release identification number, and it is labeled as unallocated by server.Then, this numeral can be reused for another calling.In order to have higher availability, server will have backup, and during with convenient server failure, backup can be taken over the work of server.Yet this can cause bottleneck and autgmentability poor.
And, for each user in the communication system, one group of data that constitute client's description (customprofile) of user is arranged.Client describes the information comprise the communication system service that is authorized to use such as address name and address, user, client's accounting information etc.When the user opens travelling carriage (such as cell phone), owing to mobile switching centre (MSC) prospective users can be made a phone call, so MSC caches to client's descriptor in the memory.The amount of information that client describes can be very big, and require a large amount of internal memories.Because a large amount of client's data of description are loaded into the needed time in the memory, when opening travelling carriage, client's data of description is loaded in the memory.The complete or collected works of client's data of description are called as customer data base (SDB).
Mobile switching centre (MSC) distributes International Mobile Subscriber Identity (IMSI) number or Temporary Mobile Station Identity (TMSI) number to travelling carriage.For activity associated with the data, mobile switching centre (MSC) distributes packet-based Temporary Mobile Station Identity (P-TMSI) number to travelling carriage.Customer data base (SDB) is associated with the identification number (IMSI, TMSI or P-TMSI) that mobile switching centre (MSC) distributes to travelling carriage.
When the user sends when calling out from travelling carriage (for example cell phone), this calling is routed to the mobile switching centre (MSC) that is used to register.Mobile switching centre (MSC) can send information and can receive information from customer data base (SDB) server to customer data base (SDB) server.Customer data base (SDB) server can receive the registration message that sends to mobile switching centre (MSC) from travelling carriage.Customer data base (SDB) server uses and distributes its Home Location Register (HLR) to register travelling carriage.Customer data base (SDB) server can also be stored the business description (serviceprofile) of each travelling carriage.Customer data base (SDB) server can also visited mobile switching center (MSC) during incoming call, to obtain interim routing number (for example, IMSI).
Prior art system is used unique user database (SDB) server that is arranged to main (primary)-backup configuration.Master-backup configuration has reduced single fault makes the inoperable danger of customer data base (SDB) server.Yet there is tangible problem in this simple master-backup configuration.The first, centralized user database (SDB) server is a bottleneck potential in the system.The performance of customer data base (SDB) server may be not enough to get caught up in the requirement to it.The second, have performance issue, that is, the access time of centralized user database (SDB) server may be too slow for real-time application.The 3rd, centralized user database (SDB) server has the internal memory restriction.The number of users that centralized user database (SDB) server can hold may be limited.When system adds more user, centralized user database (SDB) server may be difficult for expansion.At last, the relevant cost of centralized user database (SDB) server may be very high.
In addition, it is extremely complicated not interrupting existing service in traditional switching center's upgrading call processing server application program.In some prior art system, software upgrade requires to have full redundancy (duplex) hardware in switching center.Redundant component is divided into activation side and un-activation side.The requirement complex control software is managed this division (by exchange activation side and un-activation side) and is managed the processing that the two halves of system is merged into holonomic system.Redundant hardware makes the switching center of prior art increase extra cost, and the exploitation complex control software is also very expensive, and because its complexity is easy to make mistakes, is difficult to safeguard.
Thereby, need improved communication network device and service.Especially, the mobile switching centre that need have high reliability and during the high call traffic carrying capacity, be subjected to bottleneck condition influence minimum.Particularly, need distributed call processing architecture, improved trunk idle listserv architecture, improved call identification server architecture and user database server architecture improved, that in mobile switching centre and other similar switch, use.And needing can online upgrading and do not require the switching center that uses redundant hardware and complicated and expensive Control Software.
Summary of the invention
In order to solve the defective of above-mentioned prior art, main purpose of the present invention provides a kind of system and method, is used for carrying out distributed call fast and effectively in communication system working load shared group and handles and online upgrading.
According to first preferred embodiment of the invention, switch comprises: 1) Main Processor Unit, can carry out the call treatment client applications, and wherein, each call treatment client applications is related with one of a plurality of callings connections; With
2) N call applications node, can carry out the call processing server application program, wherein, on the first node of N call applications node, carry out the first call processing server application program, and be associated with the second call processing server application program that on the Section Point of N call applications node, carry out, similar, above-mentioned Section Point and first node are separated.Thereby, the first and second call processing server application programs form the first load Sharing group server application, wherein, each call treatment client applications sends to the first load Sharing group server application with the call treatment service request, and the first load Sharing group server application is selected one of first and second call processing server application programs, and carries out the call treatment service of being asked according to the load Distribution algorithm.
According to the first embodiment of the present invention, the load Distribution algorithm is the new call treatment service request of distribution between the first and second call processing server application programs in an alternating manner.
According to the first embodiment of the present invention, the load Distribution algorithm is according to the current call treatment load of the first call processing server application program and the current call treatment load of the second call processing server application program, the call treatment service request that distributes new.
Also according to the first embodiment of the present invention, on the level that the current call treatment load with the first call processing server application program maintains roughly with the current call treatment load of the second call processing server application program equates, the load Distribution algorithm new call treatment service request that distributes.
Also according to the first embodiment of the present invention, the first call processing server application program comprises the first master-backup group server application, wherein, the first master-backup group server application is included in the first main call treatment of carrying out on the first call applications node and the first backup call treatment related with the first main call treatment.
According to the first embodiment of the present invention, the state information relevant with the first main call treatment is mirrored onto on the first backup call treatment relevant with the first main call treatment.
Also according to the first embodiment of the present invention, the first backup call treatment is arranged in the first call applications node.
Also according to first embodiment of the invention, the first backup call treatment is arranged in the call applications node that separates with the first call applications node.
In the first embodiment of the present invention, the second call processing server application program comprises the second master-backup group server application, wherein, the second master-backup group server application is included in the second main call treatment of carrying out on the second call applications node and the second backup call treatment relevant with the second main call treatment.
In the first embodiment of the present invention, the state information relevant with the second main call treatment is mirrored onto on the second backup call treatment relevant with the second main call treatment.
Also in the first embodiment of the present invention, the second backup call treatment is arranged in the second call applications node.
Still in another embodiment of the present invention, the second backup call treatment is arranged in the call applications node that separates with the second call applications node.
In order to solve the above-mentioned defective of prior art, second purpose of the present invention provides a kind of controller, and it uses together with communication switchboard, is used for monitoring the user mode of the trunk line relevant with switch.Switch is that the calling that is used for handling between caller device on the related relay line and called device connects.Controller partly is efficient and the reliability that is used for improving the switch relevant with the distribution responsibility of related relay line.
Controller comprises N call applications node, it can be carried out and be used for connecting the trunkfree listserv application program of distributing some trunk lines to calling out, wherein, the first trunkfree listserv application program is carried out on the first call applications node, and relevant with the second trunkfree listserv application program of carrying out on the second call applications node that separates.The first and second trunkfree listserv application programs have formed load Sharing group server application, its be used for (i) receive the trunk line request for allocation of the call treatment carried out in the comfortable switch, (ii) select the first or second trunkfree listserv application program, distribute trunk line connect for the calling relevant with the trunk line request for allocation according to the load Distribution algorithm.
Second purpose of the present invention provides a kind of group policy, and it allows the server application program of variable number to add mapping resource group (MRG).This group policy allows the server group, and main using and the backup server group adds MRG, and the mapping of the particular server of management given trunk group is provided, so that client applications can distribute and release trunk effectively.This group policy allows to add, remove or compensate server application, such as in event of failure, so that the server configuration is transparent to client applications.When configuration change occurring, the MRG strategy allows its controller distribution server in communication network, to avoid single place fault, and the effectiveness of performance mechanism that is provided for distributing trunk line, so that make it in available server, in selected several servers, automatically redistribute trunk line group responsibility.
In order to solve the above-mentioned defective of prior art, the 3rd purpose of the present invention provides a kind of controller, it is used for connecting to the calling relevant with switch and distributes the call identification value, wherein, and caller device and the connection of the calling between the called device on the switch processes trunk line relevant with switch.According to a preferred embodiment of the invention, this controller comprises N call applications node, it is used to carry out a plurality of identification server application programs that connect distribution call identification value to calling out, wherein, in a plurality of identification server application programs first carried out on first of N call applications node, and it is relevant with second of the identification server application program of on second of N call applications node, carrying out, second of above-mentioned N call applications node is separated with first node, the first and second identification server application programs have formed load Sharing group server application, and, the new call treatment receipt of call identification request of load Sharing group server application wherein from just switch, carrying out, and select one of first and second identification server application programs, so that according to the load Distribution algorithm, connect for the new calling relevant and distribute the call identification value with the call identification request.
A third embodiment in accordance with the invention, the first identification server application assigned has the call identification value of first successive range, and the second identification server application assigned has the call identification value of second successive range, and second successive range is different from the first continuous model to be stranded.
The new call identification request of distribution between the first and second identification server application programs in an alternating manner of a third embodiment in accordance with the invention, load Distribution algorithm.
A third embodiment in accordance with the invention also, the load Distribution algorithm according to the first identification server application program when the pre-treatment load and the second identification server application program when the pre-treatment load, new call identification request distributes.
A third embodiment in accordance with the invention also, the load Distribution algorithm for the first identification server application program when the pre-treatment load maintain with the second identification server application program on the pre-treatment load level about equally, new call identification request distributes.
A third embodiment in accordance with the invention also, the first identification server application program comprises the first master-backup group server application, wherein, the first master-backup group server application be included in first principal mark carried out on the first call applications node know server application and with first principal mark know server application relevant first back up the identification server application program.
Also a third embodiment in accordance with the invention is known the relevant call status information of server application with first principal mark and is mirrored onto on the first backup identification server application program relevant with first principal mark knowledge server application.
A third embodiment in accordance with the invention also, the first backup identification server application program is arranged in the second call applications node.
In third embodiment of the invention, the first backup identification server application program is positioned on the call applications node that separates with the first call applications node.
In the third embodiment of the present invention, the second identification server application program comprises the second master-backup group server application, wherein, the second master-backup group server application be included in second principal mark carried out on the second call applications node know server application and with second principal mark know server application relevant second back up the identification server application program.
Also in the third embodiment of the present invention, the state information relevant with the second main call treatment is mirrored onto on the second backup call treatment relevant with the second main call treatment.
Also in the third embodiment of the present invention, the second backup identification server application program is arranged in the second call applications node.
Also in the third embodiment of the present invention, the second backup identification server application program is arranged in the call applications node that separates with the second call applications node.
In order to solve the above-mentioned defective of prior art, the 4th purpose of the present invention provides a kind of apparatus and method of having improved, and it is used for distributing fast and effectively and calling party database (SDB) information in communication system.
The 4th purpose of the present invention provides a kind of switch, and it can be handled in calling on many trunk lines relevant with switch, between caller device and the called device and connect.According to a preferred embodiment of the invention, switch comprises: 1) Main Processor Unit, and it can carry out call handler application, and wherein, each call handler application is relevant with one of calling connection; With
2) N call applications node, it can carry out the user database server application program, wherein, the first user database server application program is carried out on the first node of N call applications node, and it is relevant with the similar second user database server application program carried out on the Section Point of N call applications node, the Section Point of above-mentioned N call applications node separates with first node, thereby, the first and second call processing server application programs form customer data base load Sharing group server application, wherein, each call handler application program sends to customer data base load Sharing group server application with the customer data base service request, and the user database server application program is selected one of first and second user database server application programs, with according to the load Distribution algorithm, carry out institute's requesting users database service request.
A fourth embodiment in accordance with the invention, load Distribution algorithm be the new customer data base service request of distribution between the first and second user database server application programs in an alternating manner.
A fourth embodiment in accordance with the invention, the load Distribution algorithm is according to the current call treatment load of the first user database server application program and the new customer data base service request of current call treatment load Distribution of the second user database server application program.
Also according to recessed embodiment of the present invention, the load Distribution algorithm for the current call treatment load with the first user database server application program maintain with the roughly the same level of the current call treatment load of the second user database server application program on, and the customer data base service request that distributes new.
A fourth embodiment in accordance with the invention also, the first user database server application program comprises the first master-backup group server application, wherein, the first master-backup group server application is included in the first main user database server application program of carrying out on the first call applications node and the first backup subscriber database server application program relevant with the first main user database server application program.
A fourth embodiment in accordance with the invention also, the state information relevant with the first main user database server application program be mirrored to the first relevant backup subscriber database server application program of the first main user database server application program on.
A fourth embodiment in accordance with the invention also, the first backup subscriber database server application program is arranged in the first call applications node.
A fourth embodiment in accordance with the invention also, the first backup subscriber database server application program is arranged in the call applications node that separates with the first call applications node.
In the fourth embodiment of the present invention, the second user database server application program comprises the second master-backup group server application, wherein, the second master-backup group server application is included in the second main user database server application program of carrying out on the second call applications node and the second backup subscriber database server application program relevant with the second main user database server application program.
In the fourth embodiment of the present invention, the state information relevant with the second main user database server application program is mirrored onto on the second backup subscriber database server application program of being correlated with the second main user database server application program.
Also in the fourth embodiment of the present invention, the second backup subscriber database server application program is arranged in the second call applications node.
Also in the fourth embodiment of the present invention, the second backup subscriber database server application program is arranged in the call applications node that separates with the second call applications node.
In order to solve the above-mentioned defective of prior art, the 5th purpose of the present invention provides a kind of method of that use, a plurality of call processing server application programs that are used to upgrade in the switch that comprises N call applications node (CAN), wherein, each call processing server application program comprises main call treatment and the backup call treatment of carrying out on the CANs different in N CANs.According to a preferred embodiment of the invention, the method comprising the steps of: 1) receive shutdown command with the first call processing server application program of operationally upgrading, the first call processing server application program comprises in the first main call treatment of carrying out on the CAN and the first backup call treatment carried out on the 2nd CAN; 2) the first main call treatment is forbidden in the reception of response shutdown command, makes that no longer the first main call treatment on a CAN sends afterwards the call business relevant with the first call processing server application program; 3) the first backup call treatment being reassigned is the new main call treatment of the first call processing server application program, so that all call business in the future that will be relevant with the previous calling of being handled by the first call processing server application program all are sent to the first backup call treatment of being reassigned on the 2nd CAN, do not use and need not be directly acquainted with it or understand to send; 4) just move on on the different CAN if any, being correlated with the second call processing server application program and being positioned at the backup of second on CAN call treatment; With 5) in the first call processing server application program of installing on the CAN after upgrading, the first main call treatment after the upgrading of the first call processing server application program after the feasible upgrading is carried out on a CAN, and the first backup call treatment after the upgrading of the first call processing server application program after creating upgrading on the CAN.
According to a fifth embodiment of the invention, this method also comprises step: remove the first disabled main call treatment from a CAN.
According to a fifth embodiment of the invention, this method also comprises step: prevent from afterwards the call business relevant with new calling is sent on the first backup call treatment of reassigning.
Also according to a fifth embodiment of the invention, this method also comprises step: when all previous callings stop, remove the first backup call treatment of reassigning from the 2nd CAN.
Also according to a fifth embodiment of the invention, the first main call treatment after the upgrading has added the first load Sharing group server application, this first load Sharing group server application comprise with the upgrading after the similar call processing server application program of the first call processing server application program.
Also according to a fifth embodiment of the invention, the first load Sharing group server application is under the control of throttling scheme, and new call business that will be relevant with new calling sends to the first main call treatment after the upgrading.
Also according to a fifth embodiment of the invention, throttling scheme at first makes on the first main call treatment after the new call business of relatively small amount is sent to upgrading.
Also according to a fifth embodiment of the invention, throttling scheme is sent on the first main call treatment of upgrading back the new call business amount that increases gradually.
Characteristics of the present invention and technological merit have above very briefly been summarized, so that make those skilled in the art can understand hereinafter detailed description of the present invention better.Describe feature of the present invention and advantage hereinafter, this has constituted the main body of claims of the present invention.Those skilled in the art can understand, and they can easily use disclosed design and specific embodiment, revises or be designed for other structures that realize identical purpose of the present invention based on this.Those skilled in the art should know that also this equivalent structure does not deviate from the spirit and scope of the present invention of broadest form.
Before the detailed description of the present invention below, some vocabulary that best illustration is once used in the patent document and the definition of phrase: term " comprise " and " comprising " with and derivation be hard-core comprising; Term " or " included, refer to and/or; Phrase " be correlated with " with " is relevant with it " with and derivation can be comprise, be included in, with ... interconnect, comprise, be comprised in, be connected to or connected, be coupled to or be coupled, can communicate with, cooperate, interweave with it, side by side, approaching, be attached to or combine, have, have ... attribute, or the like; Any device, system or its part of at least a operation of system accused in term " controller ", this device can with hardware, firmware or software or in them at least two kinds combination realize.Should be noted that the function relevant with any specific controller can be centralized or distributed, this locality or long-range.Some vocabulary of whole patent documentation and the definition of phrase are provided, and one of ordinary skill in the art should be appreciated that if not as a rule these definition also are applied to this vocabulary that is defined and phrase existing and that use in the future.
Description of drawings
In order to understand the present invention and advantage thereof more completely, be described below in conjunction with accompanying drawing, wherein, similarly numeral is indicated similar object, wherein:
Fig. 1 has illustrated example wireless network according to an embodiment of the invention;
Fig. 2 understands example mobile switching centre according to an embodiment of the invention in more detail;
Fig. 3 has illustrated that according to principle of the present invention the use group is served the selected portion of the mobile switching centre that carries out the distributed call processing;
Fig. 4 A has illustrated according to the present invention the inner group policy class of server side of first example embodiment;
Fig. 4 B has illustrated according to the present invention client's side internal customer strategy architecture of first example embodiment;
Fig. 4 C has illustrated according to the present invention load Sharing client's side strategy inside structure of first example embodiment;
Fig. 4 D has illustrated according to the present invention client's side internal customer strategy architecture of second example embodiment;
Fig. 4 E has illustrated according to the present invention load Sharing client's side strategy internal architecture of second example embodiment;
Fig. 5 has illustrated that according to second embodiment of the invention the flow chart of second exemplary method of operation control, this controller comprise at least one load Sharing group;
Fig. 6 is the flow chart of operation that the 4th example embodiment of method of the present invention has been described;
Fig. 7 has illustrated the subregion of the call processing server application program in the mobile switching centre of the 5th example embodiment and the flow chart of online upgrading according to the present invention; With
Fig. 8 A-8K is when the call applications node being carried out the processing of described subregion of Fig. 7 and online upgrading, the series of drawing of call applications node in the example mobile switching centre (MSC).
Embodiment
The Fig. 1 that discusses below is used to describe principle of the present invention to Fig. 8 with in patent document a plurality of embodiment only are used for explanation, in no case should be counted as having limited scope of the present invention.Those skilled in the art will appreciate that principle of the present invention can realize in the suitable communication network of any configuration.
In the open text below, in the mobile switching centre of cordless communication network, realized being used to carry out the group service framework of various distributed call processing capacities.This just is used for explanation and should be considered to limit scope of the present invention.Person of skill in the art will appreciate that following group service framework can be realized, comprise multiple switch, router etc., and also comprise the whole cable network that does not comprise wireless device in the communication network of other type and device.
Fig. 1 has illustrated example wireless network 100 according to an embodiment of the invention.Wireless network 100 comprises a plurality of cell sites (cell site) 121-123, and each cell site all comprises a plurality of base stations, BS101, BS102 or BS103, in one.Base station 101-103 warp, for example, code division multiple access (CDMA) channel is communicated by letter with a plurality of travelling carriages (MS) 111-114.Travelling carriage 111-114 can be any suitable wireless device, comprises traditional cellular radio telephone, PCS hand-held device, personal digital assistant, portable computer or metering device.The invention is not restricted to mobile device.Also can use accessing terminal of other type, comprise fixed radio terminal.Yet,, hereinafter only show and the discussion travelling carriage for easy.
Dotted line has shown the approximate boundaries of the cell site 121-123 that base station 101-103 is gone out.For description and interpretation, cell site is shown as approximate circle.It should be clearly know that cell site can have other irregularly shaped, this depends on selected cell configuration and nature and artificial obstacle.
As be known in the art, cell site 121-123 is made up of a plurality of sectors (not shown), and each sector is illustrated by the directional antenna that is connected with the base station.The embodiment of Fig. 1 has illustrated the base station of center of housing estate.Alternative embodiment is placed on directional antenna in the corner of sector.System of the present invention is not limited to any cell site configuration.
In one embodiment of the invention, BS101, BS102 or BS103 comprise base station controller (BSC) and one or more BTS under CROS environment (BTS).Base station controller and BTS under CROS environment are as well known to those skilled in the art.Base station controller is a kind of device of management of wireless communications resource, comprises the base station transceiver that is used in the sub-district of cordless communication network appointment.BTS under CROS environment comprises RF transceiver, antenna and is positioned at the electric equipment of each cell site.This equipment can comprise air-conditioning unit, heating unit, power supply, telephone line interface and RF transmitter and RF receiver.In order to explain operation of the present invention simply, the BTS under CROS environment in each sub-district 121-123 is totally represented with BS101, BS102 or BS103 respectively with the base station controller relevant with each BTS under CROS environment.
BS101, BS102 or BS103 are transmitting the voice-and-data signal each other and between the PSTN (PSTN) through communication trunk 131, (MSC) 140 of mobile switching centre and communication trunk 132.Trunk line 131 also provides access path to transmit control signal between MSC140 and BS101, BS102 and BS103, this control signal is used to set up the connection that is used for the voice-and-data circuit, this connection is through the connection of communication trunk 131 between MSC140 and BS101, BS102, BS103, and through the connection of communication trunk 132 between MSC140 and the Internet or PSTN.In some embodiments of the invention, communication trunk 131 can be several different data link, and here, every data link is connected to MSC140 with one among BS101, BS102 or the BS103.
Trunk line 131 and 132 comprises one or more any suitable jockeys, and this device comprises that T1 line, T3 line, optical fiber link, network packet data backbone connect or the data of any other type connect.Those skilled in the art can know: being connected on the trunk line 131 and 132 can be provided for transportation simulator transmission of audio signals path, be used for pulse code modulated (PCM) form transmission of speech signals data path, be used for Internet protocol (IP) form transmission of speech signals digital path, be used for the digital path of ATM(Asynchronous Transfer Mode) form transmission of speech signals or with the digital path of other suitable connection transmission format protocol transmission of speech signals.Those skilled in the art can know that being connected on the trunk line 131 and 132 can be provided for the transmission path with appropriate signaling protocol transmission analog or digital control signal.
Fig. 2 understands example according to an embodiment of the invention mobile switching centre 140 in more detail.MSC140 is included in the interference networks 200 between other things.Interference networks 200 comprise exchange mechanism 205 and exchange control unit 210, and they provide the switching path between the telecommunication circuit in trunk line 131 and 132 together.MSC140 be provided in the wireless network 100 the user and such as service and cooperation between the user of the external network of PSTN or the Internet.With the similar mobile switching centre of MSC140 be as well known to those skilled in the art.
When wireless network user was opened his or her travelling carriage (for example cell phone) or fixedly accessed terminal, wireless information was just entering network through air interface informing base station travelling carriage (or fixedly access terminal).Yet, in trunk line 131-132, do not connect for voice or data service bearer circuit automatically.The voice or the data service that do not need to be connected to PSTN (PSTN) or the Internet connect, and call out (for example, dialing the number) or accesses network up to the user.
Yet, even when phone is idle, also must obtain some information (that is, user data) of user and it is stored among base station or MSC140 or the two, with checking user, collecting charging information, the spendable service of identification user, determine the ability of travelling carriage etc.Require the control signal (with respect to the voice-and-data business) of aforesaid operations also to be carried on trunk line 131 and 132.After in the memory of storage of subscriber data at MSC140, these data can be used by multiple call treatment client (CPC) application program, and when travelling carriage was activated, this call treatment client applications can be started by user or another device.
For example, when MS111 was opened earlier, for MS111 sets up call treatment, user data (for example, charge information) was stored among the MSC140 in MSC140, and MSC140 can be visited by call treatment or other call applications of the call service that particular type is provided.If the user dials the number on MS111 or receives the calling that sends to PSTN from MS111, the call treatment of MS111 is just set up to call out on one of trunk line in one of trunk line on trunk line 131 and the trunk line 132 and is connected.The MS111 call treatment of carrying out in MSC140 is safeguarded the state information that all and this calling are relevant with MS111, and handles desired all other application of MS111, comprises third-party call, voice mail, calling disconnection etc.
In order to handle a large amount of call business, the call treatment of many activation and the call service application of being handled by MS111 are necessary to distribute on a plurality of call applications nodes.Call service can comprise the application that is used for calling party database, selection (or no longer selecting) trunk line, acquiring and maintaining call identification information etc.The invention provides and be used for several different methods and the equipment used with high reliability and redundant fashion distribution call treatment and call service on a plurality of call applications nodes.This is to realize that by the distributed network of redundant server wherein, the distribution call business is in order to increase the call handling capacity of MSC140.The redundancy of distributed server all is transparent to the call treatment client applications of request service with the call processing server application program of service is provided.This has also reduced the complexity of client and server application.
Fig. 3 understands according to principle of the present invention in more detail, and the use group is served the selected portion of carrying out the example mobile switching centre 140 that distributed call handles.MSC140 comprises Main Processor Unit (MPU) 310, system administration manager node 1 (SYSMGR1), optional system administration manager node 2 (SYSMGR2) and master data base 320.MSC140 also comprises a plurality of call applications nodes (CAN) and a plurality of local storage device (SD), and the call applications node comprises CAN1, CAN2 and CAN3, SD1, SD2 and SD3 that local storage device (SD) is just relevant with CAN3 with CAN1, CAN2.Master data base 320 also can be used as the major software storehouse of stored data base, software image, server statistics, logon data etc.SD1-SD3 can be used for storing local encapsulation (local capsule), transient data etc.
System administration manager node 1 and 2 and CAN1-CAN3 carry out configuration management (CM) together and handle, it sets up each node with appropriate software and configuration data when initial start-up or after restarting.Each node in the system is also carried out the monitoring nodes (NM) of the software of packing into and is handled and determine whether to handle the tracking processing of failing, and system administration manager node 1 and 2 is carried out first and handled P1 arbitrarily, and system administration manager node 1 is also carried out second any processing P2.
According to principle of the present invention, call applications node 1-3 (CAN1-CAN3) carries out a plurality of call treatment (CP) server applications of leading usefulness and back-up processing that are organized into, it can be used as the distributed group service and is used for the 1st to N call treatment client (CPC) application, that is the CPCAPP1-CPC APPn in the Main Processor Unit 310.N call applications node (for example, CAN1-CAN3) is the computing node that separates that comprises processor and memory, provides retractility and redundancy by adding the more calling application node simply.
During N call treatment client (CPC) uses each, that is, each among the CPCAPP1-CPC APPn among the MPU310 is all handled control signal relevant with individual call and message, and this individual call is relevant with travelling carriage.Among the CPC APP1-CPC APPn each is set up upright session with load Sharing, and this load Sharing group is given master-backup group call processing server application program with call distribution, i.e. CP1, CP2 or CP3, in some.The call treatment services/functionalities that the actual execution of selected call processing server application program call treatment client applications is asked.
In the illustrated embodiment, carried out 3 exemplary call processing server application programs, that is, and CP1, CP2 and CP3.Each of these processing all exists as master-backup group.Like this, CP1 exists as main CP1 (P) of processing and back-up processing CP1 (B).Similarly, CP2 exists as main CP2 (P) of processing and back-up processing CP2 (B).CP3 exists as main CP3 (P) of processing and back-up processing CP3 (B).In the illustrated embodiment, CP1 (P) is positioned on the different call applications node (that is, CAN1 and CAN2) with CP1 (B).This is not strict requirement: CP1 (P) can be positioned on the identical call applications node (for example CAN1) with CP1 (B), and still is provided for the reliability and the redundancy of the software fault of the main CP1 of processing (P).Yet in a preferred embodiment of the invention, the main processing is positioned on the different call applications nodes with back-up processing, thereby hardware redundancy and software redundancy are provided.Like this, CP1 (P) and CP1 (B) are positioned on CAN1 and the CAN2, and CP2 (P) and CP2 (B) are positioned on CAN2 and the CAN3, and CP3 (P) and CP3 (B) are positioned on CAN3 and the CAN1.
CP1, CP2 and CP3 are formed for super group of load Sharing together.Like this, CP1 (P) and CP1 (B), CP2 (P) and CP (B) and CP3 (P) and CP3 (B) are the parts of the first load Sharing group (LSG1), and LSG1 with dashed lines border is represented.In addition, CAN1-CAN3 holds (host) 3 other load Sharing groups, that is, and and LSG2, LSG3 and LSG4.LSG2 comprises two trunkfree tabulation (TIL) server applications, that is, and and TIL1 and TIL2.TIL1 exists as main TIL1 (P) of processing on the CAN2 and the back-up processing TIL1 (B) on the CAN3.TIL2 exists as main TIL2 (P) of processing on the CAN3 and the back-up processing TIL2 (B) on the CAN2.Similarly, LSG3 comprises two identification servers (IS) end application program, that is, and and IS1 and IS2.IS1 exists as main IS1 (P) of processing on the CAN1 and the back-up processing IS1 (B) on the CAN2.IS2 exists as main IS2 (P) of processing on the CAN2 and the back-up processing IS2 (B) on the CAN1.At last, LSG4 comprises two customer data bases (SDB) server application, that is, and and SDB1 and SDB2.SDB1 exists as main treatment S DB1 (P) on the CAN2 and the back-up processing SDB1 (B) on the CAN3.SDB2 exists as main treatment S DB2 (P) on the CAN3 and the back-up processing SDB2 (B) on the CAN2.
The group service is provided for the framework of tissue one distribution type software object in computing network.Each software object provides service.In addition, the group service framework provides enhancing behavior (enhanced behavior), member that it is used for, and definite group membership concerns, what action and control group are taked in decision when appearance is wrong and the clean culture between the client, multicast, cast communication.The behavior of strategy with the service that strengthens this group and provided is provided group.Some strategies comprise master-backup, and it is used for high service availability; And load Sharing, it is the distribution service load in network.
Be arranged in the call processing server application program of computing network, the service that provides client applications to call such as CP1-CP3, IS1-IS2 and TIL1-TIL2 is such as CPC APP1-CPC APPn.As shown in Figure 3, it is master-backup group that the call processing server application program is organized into, and this master-backup group is configured to the master-backup group of 1+1 type.These master-backup groups have a plurality of members, and exact amount can be flexible according to the quantity of processing and/or employed computing node (CAN).All master-backup groups all are single load Sharing group (LSG1 for example, LSG2, LSG3, members LSG4).
Especially be noted that when the call treatment client applications when CPC APP1-CPC APPn was the client with respect to call processing server application program CP1, CP2 and CP3, server application can be the client with respect to another server application.Especially, call processing server application program CP1-CP3 is with respect to trunk idle listserv application program TIL1, TIL2, user database server application program SDB1, and SDB2 and identification server application program IS1, IS2 can be the client.
Client applications is set up the interface with the load Sharing group.When client applications was received new call instruction, client applications was just according to the session of client's lateral load sharing policy foundation with the load Sharing group.Initial server (CP1, CP2 etc.) selection strategy is based on CPU and uses (that is, based on the processor load among each candidate CAN, selecting the lighter group of load earlier), but, can use other algorithm policy, such as round-robin method (that is, in turn to the new calling of each CAN distribution).
Client applications is associated new calling with session, and sends the message relevant with this calling through this session object.Client applications also through set up and master-backup group between the session message of receive autonomous-backup group.Only will lead-main handle (for example, the CP1 (P)) of backup group add the load Sharing group (for example, LSG1).
Owing to many reasons, so, can from service, remove the application that comprises main member.Along with removing of main member, the backup member becomes main member, and server application no longer is the member of load Sharing group.Yet, the session between that client applications also is kept for existing call and master-backup group.If master-backup group is not the member of load Sharing group, just not to its new calling that distributes.
If the main member as master-backup group of load Sharing group membership breaks down, just notice backup member master member is out of order (or leaving), and, back up the role that the member takes on main member then.Must carry out the responsibility of these actions by server application.The responsibility of group service is that notice backup member master member is out of order or is left.
As the part that online software upgrade is handled, the one or more application that comprise master-backup group can be removed, be unloaded from service, and the software code with redaction recovers then.If the interface of these groups does not change, they just add existing load Sharing group.When it is activated for the first time, require customer interface to carry out throttling to call business to specific master-backup group.Professional throttling is not used from 0% (have and call out) and is represented to 100% percentage change.All the new callings according to the dispatching algorithm scheduling are handled in this session.Restriction coefficient is both initialized to 100% for the master-backup group of any adding load Sharing group.During online software upgrade, new software version is not adjusted to restriction coefficient from there not being calling.The session between the master-backup group can be set up and specify to any client applications that is used for the load Sharing group.Then, the client can change restriction coefficient at any time.When restriction coefficient changed, all client sessions interfaces are the restriction coefficient after multicast receives change all.Along with the increase of restriction coefficient, the call processing server application program of new software version can receive more substantial call business.
Call treatment communication from client applications to call processing server master-backup group must be supported very a large amount of callings.The internal transmission that the utilization of group software is made up of multicast protocol (simple IP multileaving) and optional unicast protocol.Unicast protocol is TCP/IP, SCTP or other host-host protocol.Multicast protocol is used for changing the inside member relevant with error detection with member relation, state and communicates by letter.When not having unicast transmission, multicast protocol is used to transmit the client/server communication stream.When unicast protocol is provided, when unicast protocol is used to provide high velocity stream between client and the server.Stream always is sent to the main member of master-backup group, and this master-backup group all is transparent to call treatment client applications and call treatment (for example, CP1, CP2, CP3, TIL1, TIL2, IS1, IS2).
As mentioned above, the call treatment on the call applications node (CAN) is organized into the load Sharing group.Each call treatment (for example, CP1, CP2, CP3, TIL1, TIL2, IS1, IS2) himself is exactly a master-backup group.The member of master-backup group can provide service, but has only the main member of this group to receive message, and thereby actually provides service.When a member of group is chosen as main member, be the one or more interface streams of this group registration.Each stream is the stand-alone interface that is used for some call treatment services.
Call treatment client applications among the MSC 140 (for example CPC APP1, CPC APP2) receives the new indication of calling out, and with the group service select with call applications node (that is, server) between interface, should new calling to handle.Call treatment on each server (CAN) is the member of load Sharing group, and uses (that is, load) or other algorithm (such as round-robin method) based on CPU, from the angle Selection specific call application node (CAN) of call treatment client applications.For selected specific master-backup group, session turns back to this call treatment client applications.When set up and master-backup call processing server group between session the time, then, the call treatment client applications is opened to the interface (representing the interface to master-backup group) of special member and obtains session interface.Each call processing server sends and the relevant message of new calling through session interface.Send any afterwards the affairs (transaction) relevant through identical session object with calling.
Call processing server (that is master-backup group) can be through using one or more session asynchronous transmission message that define stream interface.The main member of call processing server group receives affairs.The backup group member does not receive affairs.Main group membership sends to the backup group member with fresh information (updates).In example embodiment of the present invention, when main group membership's decision sends fresh information to the backup group member.Main member begins to send fresh information in calling when being replied.Before calling is replied, calling is defined as transient state calls out.After calling is replied, calling is defined as stable state calls out.The present invention protects stable state to call out failsafe, but does not protect transient state to call out.Yet, more early and send fresh information from main member to the backup member more continually and also can protect transient state to call out failsafe.
If main group membership is out of order, so, the backup group member becomes new main member.According to example embodiment of the present invention, all transient state call informations during (fail-over) period after the fault (main member is out of order becomes time between the new main member to the backup member) may be lost, but all stable calls information are all safeguarded by backup.
Best is, the present invention and has hidden system size to master-backup group server application, call treatment client applications and the two to the retractility of system without limits.The present invention has eliminated any single place fault in the system.Intrasystem any fault does not influence system availability and performance.
Can dynamically add new call applications node (CAN) and additional master-backup group server application (for example CP1, CP2, CP3, TIL1, TIL2, IS1, IS2), and can begin to be new call business service to the load Sharing group.The call treatment client applications is not added the influence of new server.If server is out of order, its backup just is responsible for load.This just provides high availability and has made call drop minimum for each call service.
1. first embodiment (call processing server application program)
Fig. 4 A is to the explanation according to the inner group policy class of the server side of first example embodiment of the present invention.Fig. 4 B is to the explanation according to client's side internal customer of first example embodiment of the present invention strategy architecture.Fig. 4 C is to the explanation according to load Sharing client's side strategy internal architecture of first example embodiment of the present invention.
Server application produces master-backup group, adds this master-backup group then.This move produces the server side strategy that comprises the object shown in Fig. 4 A.The group policy distribution is called from the client's, and main member's distributed selection in the participation group keeps group membership's relation, monitoring group membership's fault.Server application uses the group adapter object of acting on behalf of the member as the load Sharing group to add the load Sharing group.Before adding the load Sharing group, set group adapter object (groupadaptor objects) with the name of master-backup group.
Client applications is established to the customer interface of load Sharing group and begins by opening session.The behavior of opening session utilizes client's lateral load sharing policy to select one of member of load Sharing group.Fig. 4 C has shown that internal load shares the customer strategy architecture.Session object self has been included the client's side strategy that links to each other with specific master-backup group.Fig. 4 B has shown the internal architecture that is used for this customer strategy.
The call treatment client applications is communicated by letter with selected server (it is the main member in the master-backup group).Because main call treatment receives the message from the call treatment client applications, so main call treatment is upgraded state and sent to corresponding backup call treatment.If main call treatment is out of order, just select backup call treatment automatically as new main call treatment.During started working to new main call treatment period after the fault, the call treatment client applications received faulty indication, and can attempt again sending up to new main call treatment and prepare to receive message.This makes, and the messaging service of losing in period reaches minimum after the fault.In case by using session to finish the call treatment client applications, the call treatment client applications just can discharge session.
The PBUNIGroupPolicy group policy has following inside member:
1) PBUNIConfiguration-is designated " PBUNI " with the group policy name, and specifies the qos requirement of the communication stack that is used for this strategy.
2) member relation of PolicyGroupMembership-maintenance group, and provide member relation agreement, this agreement to be used to add new member, remove the member who has left, out of order member is labeled as " FAILED ".
3) PBPolicyEventNotification-is provided for the behavior of event notice, such as i) when member's adding group (recovery), ii) leave group (leaving), iii) fault (fault), or iv) have state to change.
4) PBMemberStateControl-is useful on the state machine that main separation is selected group membership's adding being arranged, leaving when occurring with fault.The local example of such each (local instance) determines which member is main.Might divide (partition) because of network more than one main member can be arranged at the same time.
5) PBSessionControl-is controlled at call treatment client applications and the foundation of the session between the group membership that is used for master-backup group.
6) PBPolicyIOControl-is provided for the master-backup policy behavior of multileaving and sends to the group membership.
7) GroupSendProtocol-is provided for group membership's agreement of the information that sends to the client of other member of this group and this group.
8) UnicastGroupInterface-provides the group interface of stand-alone interface for each encapsulation of the resident group membership of having wherein.
PBUNIClientPolicy is a kind of master-backup client strategy, and wherein, cast link is used for and group communication.General transmission only sends to main member and is not sent to the backup member redundantly.PBUNIClientPolicy has following member.
1) ClientMembershipView-provides the part figure that the group membership concerns, but not as GroupMembershipView, it does not add the agreement relevant with group membership.
2) PBUNIClientPolicyIO-handles the I/O of the cast link through arriving main member.
3) GroupSendProtocol-provides the group membership agreement, and it is used for sending information to the client of other member of this group and this group.
4) the ClientSessionControl-session of group membership management client side.
5) PBUNIClientStateControl-keeps local figure, and its member is the main member of master-backup group.
6) the ClientSessionControl-session of group membership management client side.
7) UnicastGroupInterface-provides stand-alone interface for each encapsulation that the group membership was positioned at.
LSIClientPolicy is that to be used for himself be exactly group membership's the load Sharing strategy of group.When the session that is opened to the member, LSIClientPolicy provides the member of client's side the selection that circulates.Each session provides group interface for particular group.LSIClientPolicy returns each session support to the message throttling is provided.Throttling can change to 100% (send all message, this message is generally selected with round-robin scheduling) from 0% (no message).LSIClientPolicy has precedence over the strategy of basic ClientPolicy.LSIClientPolicy comprises following member:
1) LSIClientNotification-is event notice ClientPolicy notification procedure (notifier) and local notification procedure.
2) LISIClientSessionControl-returns session with round-robin algorithm.The session that is provided is the interface to another group.LISISessionControl is useful on the operation index (runningindex) of selecting new session, and this new session is used for each and opens conversation request.LISISessionControl has the known interface tabulation that is called as " known member ".Known member is by the mapping table of member id index (map) and comprises the SessionCount object that it comprises the reference count of the number of users of actual session and session instance.When even the member leaves this group, also keep known session.When the member leaves this group, from available member, remove them, but it is kept among the known member.Even this allows the client when having left the load Sharing group, also can continue to use group interface.
3) GroupSendProtocol-provides the group membership agreement, and it is used for sending information to the client of other member of this group and this group.
4) GroupInterface-is an interface class to multicast and/or the unicast protocol stack that group interface uses.
5) ClientPolicyIO-is responsible for handling client I/O.
6) ClientStateControl-is used to control the state-event of this group and the state-event of this group of acquisition.
2. second embodiment (trunk idle listserv application program)
Fig. 4 A is the explanation to the inner group policy class of server side of second example embodiment according to the present invention.Fig. 4 B is the explanation to client's side internal customer strategy architecture of second example embodiment according to the present invention.Because with the description of first embodiment very elephant, so no longer last figure is described.
Fig. 4 D is the explanation to load Sharing client's side strategy internal architecture of second example embodiment according to the present invention.
MappedResourceClientPolicy is that to be used for himself be exactly group membership's the load Sharing strategy of group.When being opened to a member's session, MappedResourceClientPolicy provides the member's of client's side circulation to select.Each session provides group interface to specific group.MappedResourceClientPolicy returns each session support to the message throttling is provided.Throttling can change to 100% (send all message, this message is generally selected with round-robin scheduling) from 0% (no message).MappedResourceClientPolicy has precedence over the strategy among the basic ClientPolicy.MappedResourceClientPolicy comprises following member:
1) ClientEventNotification-gives ClientPolicy notification procedure and local notification procedure event notice.
2) ResourceClientSessionControl-returns session with round-robin algorithm.The session that is provided is the interface to another group.ResourceClientSessionControl is useful on the operation index of selecting new session, and this new session is used for each and opens conversation request.ResourceClientSessionControl has the known interface tabulation that is called as " known member ".Known member is by the mapping table of member id index and comprises the SessionCount object that it comprises the reference count of the number of users of actual session and session instance.Even when the member leaves this group, also keep known session.When the member leaves this group, from available member, remove them, but it is kept among the known member.Even this allows the client when leaving the load Sharing group, also can continue to use group interface.
3) GroupSendProtocol-is provided for group membership's agreement, is used for sending information to other member of this group and the client of this group.
4) GSInterface-is an interface class for the multicast and/or the unicast protocol stack that are used by group interface.
5) ClientPolicyIO-is responsible for handling client I/O.
6) ClientStateControl-is used to control the state-event and the state-event that obtains this group of this group.
Fig. 4 E is the explanation to the trunkfree listserv side strategy internal architecture of second example embodiment according to the present invention.ResourceGroupPolicy is a GroupPolicy and comprises ResourceSessionControl, ResourcePolicyIOControl, ResourceGroupConfiguration, PolicyGroupMembership, GroupSendProtocol, PBMemberStateControl, GSInterface and PBPolicyEventNotification.This is to be used for super group server side structure.
Trunkfree tabulation (TIL) group is super a group of member, and here, each organizes itself master-backup group (hereinafter being called " PBG member ") exactly.A PBG member who adds TIL is chosen as the leader, and responsible task: find the trunk line tabulation resource that must manage and be that each other PBG member distributes the trunkfree responsibility of tabulating.Then, the leading member of TIL group notifies the tabulation of the relaying group of the necessary management of TIL group.Along with each new PBG member adds this group, distribute the scope of the trunk line group that will manage to it.Each trunk line group comprises the trunk line of variable number.Leader PBG member notifies each new PBG member trunk line quantity in the trunk line group that each distributes.
The client applications of TIL group receives expression and calls out the incident of initiating.Calling out the given trunk of initiating through the particular trunk group arrives.Use the trunk line in the client TIL interface assignment trunk line group or it is labeled as " using ".The specific PBG member's of management resource position is transparent to the client.When associated call has been specified a callee, calculate the trunk line group, and by the TIL customer interface, the PBG member that request is responsible for distributes any trunk line in the trunk line group.When end of calling, client applications is called out the trunk line group and the trunk line of each side by specifying this, discharges the trunk line that is used to initiate and stop this calling.
In all cases, client applications does not know which PBG trunkfree listserv application program handling the trunk line group of appointment.This just allows to add at any time Additional servers.Like this, this system all is telescopic to any amount of server.Because each member all by master-backup to forming, and bring in constant renewal in the distribution and/or the release of status data in backup, so trunkfree tabulation load Sharing group is not vulnerable to the influence of single place server failure of any one PBG member in accordance with the principles of the present invention.
Fig. 5 has illustrated the flow chart (representing with 500 usually) of the exemplary method of the controller 505 of operating LSG2 according to an embodiment of the invention, as to comprise Fig. 3.The user mode of controller 505 with the monitoring trunk line relevant with switch MSC140 runs the example, and comprise CAN2 and CAN3 intuitively, each (for example can carry out one or more trunkfree listserv application programs among CAN2 and the CAN3, TIL1 and TIL2), each trunkfree listserv application program can both be given to call out to connect and be distributed a plurality of trunk lines.As previously mentioned, LSG2 comprises TIL1 and TIL2, and wherein, TIL1 is implemented as the back-up processing TIL1 (B) that handles on TIL1 (P) and the CAN3 that leads on the CAN2.TIL2 is implemented as the back-up processing TIL2 (B) that handles on TIL2 (P) and the CAN2 that leads on the CAN3.
In order to illustrate, suppose that operation MSC140 connects (beginning step to handle in caller on many trunk line 131-132 relevant with MSC140 and the calling between the called device MS111-114; Call is initiated on main line and is stopped).Main line is organized into and is the trunk line group, and the trunk line group of variable number can be arranged, and wherein, each relaying group can have the trunk line of variable number.When on the trunk line in the given trunk group during telephone calling, this trunk line suitably is labeled as in the use.For the call that will be connected by reality, same distribute and mark particular trunk group in the termination trunk line.The trunk line that the modification Status Flag is distributed to guarantee is not used to stop the call that other has been initiated.
According to illustrated embodiment, the call treatment that LSG2 carries out from MSC140 (for example, CP1-CP3) receives trunk line request for allocation (step S510).For example, the MS113 request is connected with the call of MS112, and the request of BS102 in CP1_CP3 distributed from the trunk line of LSG2, and CP1_CP3 carries out in MSC140.According to example embodiment, in order to distribute trunk line, call treatment (for example, CP_1CP3) can be specified the special stator pack that can suitably distribute related relay line wherein, or specify particular channel, obtain particular trunk to use the son group.
LSG2 responds request for allocation, selects (a step 515 in the trunkfree listserv application program; For example, TIL1 (P) or TIL2 (P)), bootup window connects trunk line is distributed to the calling relevant with the trunk line request for allocation, and these all carry out (step 520) according to suitable load Distribution algorithm.For example, LSG2 selects among the TIL1 (P) carry out respectively and the TIL2 (P) in CAN2 and CAN3, thereby it gives the calling connection distribution trunk line relevant with the trunk line request for allocation of receiving from MS113.Suppose that MS 112 is available, just through the MS113 of BS102 with between the MS112 of BS101/BS103, set up call and connect.According to example embodiment, LSG2 (that is, in TIL1 (P) or TIL2 (P)) distributes trunk line, and channel is returned call treatment (for example, CP1-CP3) with other related data.
A release trunk (step 525 in the trunkfree listserv application program is used in the termination of LSG2 response call; For example, TIL1 (P) or TIL2 (P)).According to example embodiment, LSG2 (that is, in TIL1 (P) or TIL2 (P)) is by the particular channel release trunk, and revises the trunkfree tabulation, to indicate this trunk line not in use.
The trunk idle tabulation is stored in the memory, and is preferably backed up with the trunkfree list state, so in event of failure, do not lose current distribution list.If a part of distribution list has been lost, just can suitably use known storage administration mechanism to recover the free list state information.The trunk idle that can suitably distribute in a plurality of processing tabulation with the Performance And Reliability of raising call treatment, and makes trunkfree tabulation allocative decision on mini system configuration and big system configuration retractility be arranged all.
Assign the relaying group of flexible scope to manage preferably for the distributed member server that is used for the trunkfree tabulation, and member server can suitably be configured to the scope in the relaying group that negotiation will be managed running time, (for example be used as member server, because system upgrade, division, fault etc.) or available resources (for example, the addition of relaying group, trunk line or other resource, subtraction etc.) increase, reduce or the function of other similar modification.
Recall moves each server application to keep the related relay line in its relevant relaying group and each relaying group.When server application receives the particular trunk request for allocation, with this trunk line be appointed as distribute trunk line.When server application receives request to any trunk line in the given trunk group when (being used for end call), server application optionally distributes the idle trunk in the given trunk group.Server application also receives the request that is used to discharge the given trunk in the given trunk group.
An importance of the present invention is that it provides a kind of group policy, and it allows the server application of variable number to join in the mapping resource group (MRG).This group policy allows the server group, the backup server group of advocating peace joins among the MRG, and provide the mapping of given server application program, the specified services management is specified the relaying group, the request client applications can effectively distribute and release trunk, it is transparent for client applications that this group policy makes server configuration, in such as event of failure, its allow to server application interpolation, remove, keep or compensate.When configuration change took place, the MRG strategy is redistribution trunk line group responsibility on selected available server automatically, allows its controller distribution server on communication network, locates fault to avoid single, and the effectiveness of performance mechanism that is provided for distributing trunk line.It is that this retractility can suitably " be hidden " for server and client applications not to the restriction of system's retractility on the other hand.Also not single place fault, and the fault in the system can not influence system availability and performance yet.
3. the 3rd embodiment (identification server application program)
Fig. 4 A is the explanation to the inner group policy class of server side of the 3rd example embodiment according to the present invention.Fig. 4 B is the explanation to client's side internal customer strategy architecture of the 3rd example embodiment according to the present invention.Fig. 4 C is the explanation to load Sharing client's side strategy internal architecture of the 3rd example embodiment according to the present invention.Because to its description with to the description of first embodiment very elephant, so no longer last figure is described.
According to the 3rd example embodiment of the present invention, MSC140 utilization group method of servicing provides the distribution type identification server in MSC140 and the wireless network 100.As mentioned above, IS1 and IS2 are that identification server (IS) is used, and it provides by the service such as the various client application calls of CPC APP1-CPC APPn.IS1 and IS2 server application are organized into master-backup group, and this master-backup group stretches according to employed processing and/or computing node (CANs) quantity.IS1 and IS2 server application are the load Sharing groups, i.e. the member of LSG3.Call processing server application program CP1-CP3 is the client with respect to LSG3 and IS1 and IS2 server application.Each call identification server application (for example, IS1, IS2) is the part of master-backup group server application.
In general, the present invention uses the group that N call identification server (without limits) arranged, the call identification scope of this call identification server admin in calling out whole sign scope.Usually, call identification is 14 a numeral.Like this, for example, for the group that 10 servers are arranged, 1/10th of the whole call identification scopes of each call identification server application (for example, IS1, IS2) management.When initiating new the calling, be used for the client application requests sign of LSG3 identification server load Sharing group.Then, the customer strategy that is used for identification server load Sharing group is selected the minimum server application of load according to following one or more standards: 1) circulate with client's side and select one of whole available server tabulation; Or 2) select the minimum server application of load, utilize and/or lower memory utilization such as lower CPU.
In case selected this server, server (for example, LS1) just distributes a call identification in its range of management number.Then, use this call identification number to identify this calling.When end of calling, because the fixing identification number ranges of each server admin, so can be by coming the call release identification number from the suitable server of call identification self poisoning.Then, server application can be used it by being unallocated with this labelled notation.
The identification server application program is relevant with the call distribution server application, wherein, when application program was handled in the client applications selective call among the MPU210, first effect that thing has been exactly the client of identification server group that the call handler application program is done was with distribution call identification number.
According to a preferred embodiment of the invention, available call identification value has configurable scope.When identification server load Sharing group (that is, when LSG3) starting, select IS1 or IS2 to be " leader ", and it distributes the subclass of call identification value for each available server.Each IS1 and IS2 group server application be the successive range of administer calls ident value all.
When client applications (for example, CP1, CP2, CP3) during access identities server load shared group, give resource reference table of client applications customer strategy, this resource reference table is the question blank according to call identification value scope tissue, and this call identification value is relevant with the interface that arrives particular server.When the server configuration change, the automatic updating form content of LSG3 identification server load Sharing group.
When calling out beginning, distribute call identification at customer interface.The LSG3 customer strategy receives load factor from IS1 and IS2 server.When distributing new call identification, the LSG3 customer strategy is considered the load of server and is selected the minimum server of load.As long as each client applications has the call identification value, any client applications in the network can read or write related data now.Client applications designated call sign, and turn back to and IS1 or IS2 server between interface.Then, client applications can be to reading or writing related data or call release ident value.
4. the 4th embodiment (user database server)
Fig. 4 A is the explanation to the inner group policy class of server side of the 4th example embodiment according to the present invention.Fig. 4 B is the explanation to client's side internal customer strategy architecture of the 4th example embodiment according to the present invention.Fig. 4 C is the explanation to load Sharing client's side strategy internal architecture of the 4th example embodiment according to the present invention.Because to its description with to the description of first embodiment very elephant, so no longer last figure is described.
For each user in the communication network, one group of data that constitute client's description of user is arranged.Client describes and comprises information, such as, the communication system service that address name and address, user are authorized to use, client's charge information etc.When the user opens travelling carriage (such as cell phone), owing to mobile switching centre (MSC) 140 prospective users can be made a phone call, so MSC caches to client's descriptor in the memory.The amount of information that client describes can be very big, and require a large amount of internal memories.Because a large amount of client's data of description are loaded into the needed time in the memory, client's data of description is loaded in the memory when opening travelling carriage.The complete or collected works that client describes are called as customer data base (SDB).
(MSC) 140 of mobile switching centre distributes International Mobile Subscriber Identity (IMSI) number or Temporary Mobile Station Identity (TMSI) number to travelling carriage.For the activity relevant with data, (MSC) 140 of mobile switching centre assigns packet-based Temporary Mobile Station Identity (P-TMSI) number to travelling carriage.Customer data base (SDB) is associated with the identification number (IMSI, TMSI or P-TMSI) that mobile switching centre (MSC) 140 distributes to travelling carriage.Following elaboration is relevant with IMSI, also relates to TMSI number and P-TMSI number.
When the user sends when calling out from travelling carriage (for example cell phone), this calling is routed to the mobile switching centre (MSC) 140 that is used to register.Mobile switching centre (MSC) 140 can send information and can receive information by customer data base (SDB) server application from LSG4 by customer data base (SDB) server application in LSG4.Customer data base among the LSD4 (SDB) server application, SDB1 and SDB2 can receive the registration message that sends to MSC140 from travelling carriage.Customer data base (SDB) server application, SDB1 and SDB2, in each use Home Location Register (HLR) (not shown) distributed to register travelling carriage.Customer data base (SDB) server application, SDB1 and SDB2 also can both store the business description that is used for each travelling carriage.Customer data base (SDB) server application, SDB1 and SDB2 also can both visit MSC140, to obtain interim routing number (for example, IMSI) during incoming call.
Prior art system is used unique user database (SDB) server that is arranged master-backup configuration.Master-backup configuration has reduced single fault makes the inoperable danger of customer data base (SDB).Yet there is tangible problem in this simple master-backup configuration.The first, centralized user database (SDB) server is a bottleneck potential in the system.The performance of customer data base (SDB) server may be not enough to get caught up in the requirement to it.The second, have performance issue, that is, the access time of centralized user database (SDB) server may be too slow for real-time application.The 3rd, centralized user database (SDB) server has the internal memory restriction.The number of users that centralized user database (SDB) server can hold may be limited.When system adds more user, centralized user database (SDB) server may be difficult for expansion.At last, the relevant cost of centralized user database (SDB) server may be very high.
The present invention distributes fast and effectively by use group service framework and calling party database (SDB) information, has solved these problems of prior art systems.This group service framework provides distribution type customer data base (SDB) server.Especially, mobile switching centre (MSC) 140 comprises load Sharing group LSG4.Load Sharing group LSG4 comprises customer data base (SDB) server application SDB1 and customer data base (SDB) server application SDB2.SDB1 comprises main SDB1 (P) and backup SDB1 (B).Similarly, SDB2 comprises main SDB2 (P) and backup SDB2 (B).Main SDB1 (P) and main SDB2 (P) have formed super group of SDB.Give each SDB server application, SDB1 and SDB2 have distributed IMSI number successive range.
When call handler application program (CPA) received the calling with new IMSI number, call handler application program (CPA) was opened assigned I MSI number session.Suitable group services client strategy uses and calculates for this IMSI number and specific user's database (SDB) server application (for example, SDB1) relevant index.Group services client strategy sends to call handler application program (CPA) with this index, to give this call handler application program (CPA) interface to suitable customer data base (SDB) server application.Then, call handler application program (CPA) can write or reading of data in appropriate users database (SDB) user entry.
Though the preferred embodiment only comprises two customer data bases (SDB) server application, SDB1 and SDB2,, the invention is not restricted to use two customer data bases (SDB) server application.To a quantity part, customer data base (SDB) server application that can be used as customer data base (SDB) group without limits or constraint.
Each main customer data base (SDB) server application all has backup customer data base (SDB) server application.Customer data base (SDB) group service framework provides state to upgrade can for backup customer data base (SDB) server application.If main customer data base (SDB) server application fault, the present invention just detects this fault, and automatically switches to new main customer data base (SDB) server application (it is preceding backup customer data base (SDB) server application).It is transparent that this fault is eliminated call handler application program (CPA) or other client applications.
Like this, the invention provides a kind of telescopic equipment and method that is used at communication network operation distributed user database (SDB).The not single place of customer data base of the present invention (SDB) fault.Customer data base of the present invention (SDB) does not allow to take place performance bottleneck yet.
Fig. 6 has illustrated the flow chart of the step of the 4th preferred embodiment of operating method of the present invention.The step of this method is represented with reference number 600 usually.All distribute a successive range (step 605) of IMSI number in a plurality of customer data bases (SDB) server application each.Call handler application program (CPA) receives has new IMSI number calling, and opens assigned I MSI number session (step 610).Suitable group services client strategy uses and calculates and the relevant index (step 615) of specific user's database (SDB) server application for IMSI number.Then, group services client strategy sends to call handler application program (CPA) with this index, so that this call handler application program (CPA) of giving provides an interface (step 620) to appropriate users database (SDB) server application.
Then, mobile switching centre (MSC) 140 determines main customer data base (SDB) server applications whether online (determining step 625).If main customer data base (SDB) server application is online, so, call handler application program (CPA) can write or reading of data (step 630) in appropriate users database (SDB) user entry.
If main customer data base (SDB) server application do not have online (promptly, main customer data base (SDB) server application fault), so, mobile switching centre (MSC) 140 just uses backup customer data base (SDB) server application (step 635).Then, call handler application program (CPA) can write or reading of data (step 630) in appropriate users database (SDB) user entry.
5. the 5th embodiment (online upgrading of call processing software)
Call applications node 1-3 (promptly, the use of the master in redundant architecture CAN1-CAN3) and the mobile switching centre 140-backup group server application preferably provides a kind of distinctive method, the call processing server application program of its MSC140 that is used for upgrading and do not interrupt existing service.According to a fifth embodiment of the invention, for the division that makes the target call application node comes into force, can progressively close each the master-backup group server application on each CAN1-CAN3.Then, the target call application node can be upgraded and be new master-backup group server application software, and the software after the upgrading can progressively be reached the standard grade and join in the load Sharing group of using throttling scheme.In case the software after the upgrading is tested intact and enter in the running completely, just continues to repeat this processing on next target call application node, up to the call applications node of having upgraded all.
Fig. 7 has described flow chart 700, and it has illustrated the division and the online upgrading of the master-backup group server application in the mobile switching centre 140 in accordance with the principles of the present invention.At first, system administration manager node 1 can be specified the first target call application node (for example, CAN1) (treatment step 705) that will be upgraded automatically (or the attendant is manual).Forbid each main call treatment CPx (P) of the master-backup group call processing server application program on the first target call application node, and (for example, the respective backup call treatment CPx (B) on CAN2) becomes new main call treatment at different call applications nodes.At this moment, new main call treatment begins not have the operation of back-up processing.Yet, send custom calling service no longer for new main call treatment.Like this, CPx master-backup group just finally is closed (treatment step 710) with the disconnection of existing call.
Since the first target call application node can hold with other call applications node on the relevant one or more backup call treatment of main call treatment carried out, next the present invention backs up call treatment CPy (B) with on the first target call application node all and moves to (treatment step 415) on the different call applications nodes.The first target call application node is now without any the backup call treatment of advocating peace.The first target call application node is new division now, and all the other call applications are old division parts.
Below, the upgrade software that is used for main call treatment CPx (P) * is installed, and on the first target call application node, is produced backup call treatment CPx (B) *.Then, this new master-backup group call processing server application program join suitable load Sharing group (that is, LSG1) in.At this moment, new master server application program does not receive business.Then, use by the throttling scheme of skilled worker's (or pass through algorithm by system operator) control the increment of new call business is sent to main call treatment CPx (P) * after the upgrading, the master after upgrading-backup group CPx*100% ground moves (treatment step 720).
Then, to the second target call application node (for example, CAN2) repeating step 705,710,715,720, so that main call treatment CPz (P) * after (or generation) upgrading and backup call treatment CPz (B) * after the upgrading are installed, and operation (treatment step 725) on the second target call application node.The second target call application node and the first target call application node are newly to divide part now.
At last, the position of management software applications exchange backup call treatment CPx (B) * and CPz (B) * is so that advocate peace the not operation (treatment step 730) on identical call applications node of backup call treatment.Then, upgrading processing continues to carry out on other call applications node, all added new division up to all the other call applications nodes, and old division (comprising old software) exists no longer.
Fig. 8 A-8K is when the call applications node carries out division described in Fig. 7 and online upgrading processing, the series of drawing of call applications node in the example mobile switching centre (MSC) 140.
Fig. 8 A has illustrated the initial graph of CAN1-CAN3 in the mobile switching centre 140.
Among Fig. 8 B, the main call treatment CP1 (P) among the CAN1 stops, and relevant backup call treatment CP1 (B) has become new main call treatment CP1 (P) among the CAN2.The CP1 in CAN2 (P) does not send new business.And backup call treatment CP3 (B) has moved on to CAN2.
In Fig. 8 C, the new main call treatment CP1 of upgrading (P) * has been installed among the CAN1, has produced the new backup of upgrading call treatment CP1 (B) * in CAN1.New call business can be sent to main call treatment CP1 (P) * in the increment now, complete up to the new master of upgrading-backup group call processing server application program CP1* function.
Among Fig. 8 D, the old call treatment CP1 (P) among the CAN2 is closed at last by stopping all existing calls.
Among Fig. 8 E, the main call treatment CP2 (P) among the CAN2 stops, and the relevant backup call processing CP2 (B) among the CAN3 has become new main call treatment CP2 (P).The CP2 in CAN3 (P) does not send new business.And backup call treatment CP3 (P) has moved to CAN3.
In Fig. 8 F, the new main call treatment CP2 of upgrading (P) * has been installed in CAN2, and in CAN2, has produced the new backup of upgrading call treatment CP2 (B) *.New call business can be sent to main call treatment CP2 (P) * in the increment now, complete up to the new master who has upgraded-backup group call processing server application program CP2* function.
In Fig. 8 G, call treatment CP2 (P) old among the CAN3 is closed at last by stopping all existing calls.
In Fig. 8 H, the position among backup call treatment CP1 (B) * and CP2 (B) * exchange CAN1 and the CAN2.
In Fig. 8 I, the main call treatment CP3 (P) among the CAN3 is badly in need of new calling by leaving the load Sharing group.Main call treatment CP3 (P) and relevant back-up processing CP3 (B) thereof have stopped after all existing call service ending.
In Fig. 8 J, the new main call treatment CP3 of upgrading (P) * has been installed in CAN3, and in CAN3, has produced the new backup of upgrading call treatment CP3 (B) *.New call business can be sent to main call treatment CP3 (P) * in the increment now, complete up to the new master of upgrading-backup group call processing server application program CP3* function.
In Fig. 8 K, backup call treatment CP1 (B) *, CP2 (B) * and the position of CP3 (B) * in CAN1, CAN2 and CAN3 have been rotated, with the initial configuration that realizes illustrating among Fig. 8 A.
Though described the present invention in detail,, those skilled in the art should be known in and can carry out a plurality of changes, replacement or variation therein and do not deviate from the spirit and scope of the present invention on broad form.

Claims (10)

1. a switch can be handled caller device and the connection of the calling between the called device on many trunk lines relevant with described switch, and described switch comprises:
Main Processor Unit can be carried out the call treatment client applications, and wherein, each described call treatment client applications is relevant with one of described calling connection; With
N call applications node; Can carry out the call processing server application program; Wherein, The first call processing server application program is carried out at the first call applications node of described N call applications node; Relevant with the similar second call processing server application program of carrying out at the second call applications node of described N call applications node; The described second call applications node and the described first call applications node are separated; Thereby; The described first and second call processing server application programs form the first load shared group server application
Wherein, described each call treatment client applications sends to the described first load Sharing group server application with the call treatment service request, the described first load Sharing group server application is selected in the described first and second call processing server application programs, to carry out described call treatment service of being asked according to the load Distribution algorithm.
2. switch according to claim 1, wherein, described load Distribution algorithm is the new call treatment service request of distribution between the described first and second call processing server application programs in an alternating manner.
3. switch according to claim 1, wherein, described load Distribution algorithm is according to the current call treatment load of the described first call processing server application program and the current call treatment load of the described second call processing server application program, the call treatment service request that distributes new.
4. switch according to claim 3, wherein, the described load Distribution algorithm described new call treatment service request that distributes is so that the described current call treatment load of the described first call processing server application program maintains on the described current call treatment load level about equally with the described second call processing server application program.
5. switch according to claim 1, wherein, the described first call processing server application program comprises the first master-backup group server application, wherein, the described first master-backup group server application is included in the first main call treatment of carrying out on the described first call applications node and the first backup call treatment relevant with the described first main call treatment.
6. switch according to claim 5, wherein, the state information relevant with the described first main call treatment is mirrored onto on the described first backup call treatment relevant with the described first main call treatment.
7. switch according to claim 6, wherein, the described first backup call treatment is arranged in the described first call applications node.
8. switch according to claim 6, wherein, the described first backup call treatment is arranged in the call applications node that is separated with the described first call applications node.
9. switch according to claim 1, wherein, the described second call processing server application program comprises the second master-backup group server application, wherein, the described second master-backup group server application is included in the second main call treatment of carrying out on the described second call applications node and the second backup call treatment relevant with the described second main call treatment.
10. switch according to claim 9, wherein, the state information relevant with the described second main call treatment is mirrored onto on the described second backup call treatment relevant with the described second main call treatment.
CN 02159599 2001-12-31 2002-12-31 The system that is used for distributed call processing and online upgrading Expired - Fee Related CN100548072C (en)

Applications Claiming Priority (12)

Application Number Priority Date Filing Date Title
US10/038,879 US6917819B2 (en) 2001-12-31 2001-12-31 System and method for providing a subscriber database using group services in a telecommunication system
US10/039,186 US6862453B2 (en) 2001-12-31 2001-12-31 System and method for distributed call processing using a distributed trunk idle list
US10/038,878 US6947752B2 (en) 2001-12-31 2001-12-31 System and method for distributed call processing using load sharing groups
US10/038,872 US7366521B2 (en) 2001-12-31 2001-12-31 Distributed identity server for use in a telecommunication switch
US10/038,879 2001-12-31
US10/038,872 2001-12-31
US10/039,186 2001-12-31
US10/038,878 2001-12-31
US10/100,494 2002-03-18
US10/100,494 US20030176163A1 (en) 2002-03-18 2002-03-18 System and method for on-line upgrade of call processing software using load sharing groups
US10/174,338 2002-06-18
US10/174,338 US7308491B2 (en) 2002-03-18 2002-06-18 System and method for on-line upgrade of call processing software using group services in a telecommunication system

Publications (2)

Publication Number Publication Date
CN1434648A CN1434648A (en) 2003-08-06
CN100548072C true CN100548072C (en) 2009-10-07

Family

ID=27671290

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 02159599 Expired - Fee Related CN100548072C (en) 2001-12-31 2002-12-31 The system that is used for distributed call processing and online upgrading

Country Status (2)

Country Link
KR (1) KR100464350B1 (en)
CN (1) CN100548072C (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100983133B1 (en) * 2004-02-04 2010-09-20 엘지에릭슨 주식회사 Method for balancing load of application server in softswitch
CN100574308C (en) * 2005-05-12 2009-12-23 中国科学院计算技术研究所 Remote-apparatus access method in a kind of multi-node intelligent network application service system
KR100785393B1 (en) 2006-11-08 2007-12-13 에스케이 텔레콤주식회사 Conversion system for bypassing a call and method thereof
KR100778340B1 (en) * 2006-12-07 2007-11-22 한국전자통신연구원 Management of forwarding information table by software upgrading operation in a router system
CN102395124B (en) * 2008-06-07 2014-02-19 华为技术有限公司 Call switching method, call establishment method, call switching device and call establishment device
EP2426605B1 (en) * 2008-08-08 2016-03-02 Amazon Technologies, Inc. Providing executing programs with reliable access to non-local block data storage

Also Published As

Publication number Publication date
CN1434648A (en) 2003-08-06
KR20030058927A (en) 2003-07-07
KR100464350B1 (en) 2005-01-03

Similar Documents

Publication Publication Date Title
US6917819B2 (en) System and method for providing a subscriber database using group services in a telecommunication system
US7356577B2 (en) System and method for providing an online software upgrade in load sharing servers
CN1894985B (en) Control decisions in a communication system
CN101019384B (en) System and method for allocating and distributing end user information in a network environment
CA2495093A1 (en) Providing routing information in a communication system
CN1653738A (en) Technique for ip communication among wireless devices
WO2001039469A1 (en) A method of supplying information to a user of a mobile station
US7308491B2 (en) System and method for on-line upgrade of call processing software using group services in a telecommunication system
US20010056523A1 (en) Method and system for dynamic allocation and management of subscriber identity module memory in a telecommunication system
US20040228310A1 (en) System and method for providing an online software upgrade
US6862453B2 (en) System and method for distributed call processing using a distributed trunk idle list
US7995573B2 (en) Method and system for mobile network nodes in heterogeneous networks
CN100548072C (en) The system that is used for distributed call processing and online upgrading
CN103109502A (en) Method and apparatus of load balancing femtocell cluster access
CN103248702B (en) The server Adaptable System and method of a kind of backup ＆ disaster recovery
US8045974B2 (en) Method and system for mobile IP-nodes in heterogeneous networks
WO2000042791A1 (en) Wirelesss network and methods to be used therein
CN104272789A (en) Mobile communication system, call processing node, and communication control method
US6947752B2 (en) System and method for distributed call processing using load sharing groups
US20030176163A1 (en) System and method for on-line upgrade of call processing software using load sharing groups
CN109981795A (en) Resource request dispatching method and device
CN101997860B (en) Method and device for communication link detection management in NGN network architecture
CN1142686C (en) Method and arragement for improving service of intelligent telephone network in wireless environment
US7366521B2 (en) Distributed identity server for use in a telecommunication switch
KR100918694B1 (en) System of subscriber database server and core network in a wireless communication system

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
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20091007

Termination date: 20161231

CF01 Termination of patent right due to non-payment of annual fee