CN101119359A - Policy based service load balancing method - Google Patents

Policy based service load balancing method Download PDF

Info

Publication number
CN101119359A
CN101119359A CNA2006101039837A CN200610103983A CN101119359A CN 101119359 A CN101119359 A CN 101119359A CN A2006101039837 A CNA2006101039837 A CN A2006101039837A CN 200610103983 A CN200610103983 A CN 200610103983A CN 101119359 A CN101119359 A CN 101119359A
Authority
CN
China
Prior art keywords
server
service
channel
user
strategy
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.)
Pending
Application number
CNA2006101039837A
Other languages
Chinese (zh)
Inventor
印和平
王印龙
郭亮
胡捷
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.)
ZTE Corp
Original Assignee
ZTE Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by ZTE Corp filed Critical ZTE Corp
Priority to CNA2006101039837A priority Critical patent/CN101119359A/en
Publication of CN101119359A publication Critical patent/CN101119359A/en
Pending legal-status Critical Current

Links

Abstract

The invention discloses a server load balance method based on the strategy, solving the problem that the prior content transport network is incapable of knowing the differences of servers, learning the current statuses of the servers, and leading the customer to the equipment which has a lighter load. The server load balance method based on the strategy includes the steps as follows: (1) the turn process records the load condition of each server; (2) when the customer demands to the service, the invention chooses the server which has the lightest load serving for the customer. The present invention provides a high efficiency load balance method and has a low cost, making the whole loads in the node are distributed uniformly, achieving a real load balance.

Description

Service load balancing method based on strategy
Technical field
The present invention relates to content distribution network (CDN), relate in particular to a kind of service load balancing method based on strategy.
Background technology
Content distribution network CDN optimizes distribution with content in network, make the most popular distribution of content from the nearest fringe node of user.When user's on-demand content, overall redirector preferentially uses that nearest and substantial node provides service as the user from the user.A general node is wanted to support several thousand users, streaming media server is except providing the Video service for the user, also to bear drop-down such as content, on push away, a lot of tasks such as recorded broadcast, the service ability of convection media server has proposed very high requirement like this, therefore in order to address this problem, consider with many streaming servers to be that the user serves, the visit that is about to the user is distributed on the multiple servers, not only increased the number of users that this node can be served, and made that the load of every station server is lighter relatively.
Current, addressed the above problem a kind of DNS load-balancing method, this method is to same domain name mapping the time, can obtain a plurality of IP address, it is corresponding multiple servers, it will give a plurality of servers with independent IP address to the request of same domain name, simple, the easy row of this method, but shortcoming also clearly, it can't learn the difference between the server, can not recognize the server current state, user capture can't be directed on the lighter equipment of present load, and require operator's application domain name.Here the load-balancing method that propose solves these problems effectively.
Summary of the invention
Defective and deficiency at prior art exists the invention provides a kind of high efficiency service load balancing method based on strategy that user capture can be directed on the lighter server of load.
In order to reach the foregoing invention purpose, the present invention is based on the service load balancing method of strategy, may further comprise the steps:
(1) the master control process writes down the load condition of every station server;
(2) during user's order program service, select the lightest server of load to serve for the user.
In the above-mentioned service load balancing method based on strategy, described step (1) comprising:
(11) the master control process is recorded in the performance information of the every station server of intranodal in the equipment performance log, the channel information of intranodal is recorded in the channel table, and be that the session information of every station server distributes list item at the session information table;
(12) state information of every station server is upgraded in timing.
In the above-mentioned service load balancing method based on strategy, step (12) comprising:
(121) equipment performance log regularly upgrades the performance information of every station server;
(122) timing scan session information table, deletion surpasses the still inactive session information of set-point;
(123) timing scan equipment performance log, it is unavailable not receive that yet the server of capability message is changed to above set-point.
In the above-mentioned service load balancing method based on strategy, when intranodal has server to add or deletes, reconstructing apparatus performance table and session information table; When channel interpolation or deletion are arranged, rebuild channel table.
In the above-mentioned service load balancing method based on strategy, described step (2) comprising:
(21) if the service of user's demanding channel enters step (22); If the service of user's on-demand content enters step (23);
(22) go out channel ID according to the RTSP request analysis, travel through channel table, find the available server of this channel correspondence, enter step (23);
(23) traversal equipment performance log selects the lightest available server of load to be user's service;
(24) session information of this this service of server of record in the session information table upgrades the active user number of this server in equipment performance log;
(25) after service finished, the session information of this this service of server of deletion upgraded the active user number of this server in equipment performance log in the session information table.
In the above-mentioned service load balancing method based on strategy, step (23) comprising:
(231) traversal equipment performance log, the lightest available server of selected load;
(232) generate at the new URL that selectes server, and this URL and redirected conditional code are returned to the user;
(233) user and selected server connect, and selected server provides order program service for the user.
In the above-mentioned service load balancing method based on strategy, performance information comprises device id, Device IP, active user's number, current C PU occupancy; Channel information comprises channel ID, channel place equipment, channel map pattern; List item comprises device id and points to the pointer of Hash table.
In the above-mentioned service load balancing method based on strategy, described load is the lightest to be meant that number of users, equipment CPU usage that this server serving are minimum.
The present invention is based on the service load balancing method of strategy, dynamically determine the optimal service device by the performance information and the current sessions information of inquiry streaming media server; Can reach by running status, the load state that directly obtains underlying device and obtain current information on services, for load balancing provides facility information and information on services accurately by the inquiry session information table; Do not need the support of expensive L4/7 layer switch local load balancing.Therefore, the invention provides a kind of high efficiency load-balancing method, and cost is low, the intranodal overall load is evenly distributed, reach real load balancing.
Description of drawings
Fig. 1 is a load balancing flow chart of the present invention;
Fig. 2 is the session information list structure figure that load balancing master control process of the present invention is used.
Embodiment
The present invention is described in further detail below in conjunction with accompanying drawing:
Main points of the present invention are:
The master control process is safeguarded an equipment performance log.
The master control process is safeguarded a channel table.
The master control process is safeguarded a session information table.
The strategy of Dynamic Selection optimal device is to take all factors into consideration the performance number that information calculations such as the current number of users of serving of equipment, equipment CPU usage go out this equipment.For content-on-demand, the traversal equipment performance log filters out unavailable device, and remaining available and equipment that performance number is minimum promptly is optimal device; For channel order, need to go out channel ID according to the RTSP request analysis, travel through channel table then, calculate the performance number of channel place equipment, available and equipment that performance number is minimum promptly is optimal device.
The method applied in the present invention is:
A, master control process are safeguarded an equipment performance (containing state) table, and equipment performance log is an one-dimension array, has stored information such as device id, Device IP, active user's number, current C PU occupancy.During initialization, for every streaming media server of this node carry distributes a list item, state is changed to unavailable, sends out the performance information of the every station server of information query then, receives that the performance queries result inserts the corresponding apparatus list item; When having equipment to add on the node or deleting, reconstructing apparatus performance table.
B, streaming media server regularly report performance information, search equipment performance log according to the device id in the performance reporting information, find corresponding list item, and performance information is inserted.The scan session information table is deleted above the still inactive session information of set-point simultaneously, and the timing scan equipment performance log does not receive that for surpassing set-point the equipment of capability message is changed to its state unavailable.
C, master control process are safeguarded a channel table, and the channel information that the minute book node is all comprises channel ID, channel place equipment, channel map pattern information such as (clean culture or multicasts); When having channel to add on the node or deleting, rebuild channel table.
D, master control process are safeguarded a session information table, write down all movable session informations.The structure of session information table as shown in Figure 2, one of each hold facility, the content of list item is device id and the pointer that points to Hash table.During initialization,, distribute a Hash table to be used to store the session information of all services that this server provides simultaneously for every streaming media server of this node carry distributes a list item in the session information table.During the service beginning, find the pairing Hash table of server, insert session information in Hash table, the active user's number in the performance of the updating the equipment table when service finishes, is deleted session information, the active user's number in the performance of the updating the equipment table from Hash table.When interpolation or sweep equipment, rebuild the session information table.For new oil (gas) filling device, be this devices allocation one Hash table, the equipment for deletion finishes all session informations of this equipment, destroys Hash table then.
The strategy of E, Dynamic Selection optimal device is to take all factors into consideration the performance number that information calculations such as the current number of users of serving of equipment, equipment CPU usage go out this equipment.For content-on-demand, the traversal equipment performance log filters out unavailable device, and remaining available and equipment that performance number is minimum promptly is optimal device; For channel order, need to go out channel ID according to the RTSP request analysis, travel through channel table then, calculate the performance number of channel place equipment, available and equipment that performance number is minimum promptly is optimal device.
F, receive RTSP request at every turn, dynamically select optimal device according to the described strategy of E, generate new URL according to URL assembly unit rule, return new URL and be redirected conditional code and give client, this load-balancing session finishes.
Fig. 1 is the working timing figure of load balancing.Introduce its flow process below:
1, client (in the IPTV system, generally being set-top box) is initiated the RTSP request to the master control process.
2, the master control process accepts request, beginning load balancing flow process.
3, elementary tactics is to take all factors into consideration information calculations such as equipment active user number, CPU usage to go out this equipment performance value, determines equipment present load state according to this performance number.
4, program request divides two kinds: content-on-demand and channel order, flow process are slightly different, introduce respectively below:
1., for content-on-demand, the traversal equipment performance log, filter out unavailable device, at the equipment of available devices Dynamic Selection performance number minimum, this equipment is optimal device.
2., for channel order, go out channel ID according to the RTSP request analysis, travel through channel table then, the equipment of Dynamic Selection performance number minimum in the equipment of channel place, this equipment is optimal device.
5, obtain optimal device after, according to the rules URL assembly unit rule assembly unit URL.
6, newly-generated URL and redirected result code are returned to client.
The structure of session information table as shown in Figure 2, one of each hold facility, the content of list item is device id and the pointer that points to Hash table.During initialization,, distribute a Hash table to be used to store the session information of all services that this server provides simultaneously for every streaming media server of this node carry distributes a list item in the session information table.During the service beginning, find the pairing Hash table of server, insert session information in Hash table, the active user's number in the performance of the updating the equipment table when service finishes, is deleted session information, the active user's number in the performance of the updating the equipment table from Hash table.When interpolation or sweep equipment, rebuild the session information table.For new oil (gas) filling device, be this devices allocation one Hash table, the equipment for deletion finishes all session informations of this equipment, destroys Hash table then.

Claims (8)

1. service load balancing method based on strategy is characterized in that: may further comprise the steps:
(1) the master control process writes down the load condition of every station server;
(2) during user's order program service, select the lightest server of load to serve for the user.
2. the service load balancing method based on strategy as claimed in claim 1 is characterized in that: described step (1) comprising:
(11) the master control process is recorded in the performance information of the every station server of intranodal in the equipment performance log, the channel information of intranodal is recorded in the channel table, and be that the session information of every station server distributes list item at the session information table;
(12) state information of every station server is upgraded in timing.
3. the service load balancing method based on strategy as claimed in claim 2 is characterized in that: step (12) comprising:
(121) equipment performance log regularly upgrades the performance information of every station server;
(122) timing scan session information table, deletion surpasses the still inactive session information of set-point;
(123) timing scan equipment performance log, it is unavailable not receive that yet the server of capability message is changed to above set-point.
4. the service load balancing method based on strategy as claimed in claim 2 is characterized in that: when intranodal has server to add or deletes, and reconstructing apparatus performance table and session information table; When channel interpolation or deletion are arranged, rebuild channel table.
5. as claim 1,2 or 3 described service load balancing methods based on strategy, it is characterized in that: described step (2) comprising:
(21) if the service of user's demanding channel enters step (22); If the service of user's on-demand content enters step (23);
(22) go out channel ID according to the RTSP request analysis, travel through channel table, find the available server of this channel correspondence, enter step (23);
(23) traversal equipment performance log selects the lightest available server of load to be user's service;
(24) session information of this this service of server of record in the session information table upgrades the active user number of this server in equipment performance log;
(25) after service finished, the session information of this this service of server of deletion upgraded the active user number of this server in equipment performance log in the session information table.
6. the service load balancing method based on strategy as claimed in claim 5 is characterized in that: step (23) comprising:
(231) traversal equipment performance log, the lightest available server of selected load;
(232) generate at the new URL that selectes server, and this URL and redirected conditional code are returned to the user;
(233) user and selected server connect, and selected server provides order program service for the user.
7. the service load balancing method based on strategy as claimed in claim 2 is characterized in that: performance information comprises device id, Device IP, active user's number, current C PU occupancy; Channel information comprises channel ID, channel place equipment, channel map pattern; List item comprises device id and points to the pointer of Hash table.
8. the service load balancing method based on strategy as claimed in claim 1 is characterized in that: described load is the lightest to be meant that number of users, equipment CPU usage that this server serving are minimum.
CNA2006101039837A 2006-08-01 2006-08-01 Policy based service load balancing method Pending CN101119359A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNA2006101039837A CN101119359A (en) 2006-08-01 2006-08-01 Policy based service load balancing method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNA2006101039837A CN101119359A (en) 2006-08-01 2006-08-01 Policy based service load balancing method

Publications (1)

Publication Number Publication Date
CN101119359A true CN101119359A (en) 2008-02-06

Family

ID=39055294

Family Applications (1)

Application Number Title Priority Date Filing Date
CNA2006101039837A Pending CN101119359A (en) 2006-08-01 2006-08-01 Policy based service load balancing method

Country Status (1)

Country Link
CN (1) CN101119359A (en)

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101399965B (en) * 2008-09-28 2011-05-11 中兴通讯股份有限公司 Method and device for implementing media content positioning based on EPG server
CN102118433A (en) * 2010-12-27 2011-07-06 网宿科技股份有限公司 Multiple-tier distributed cluster system
CN102404229A (en) * 2011-12-14 2012-04-04 华为技术有限公司 System, device and method for load balancing
CN102457904A (en) * 2010-10-14 2012-05-16 深圳市泰海科技有限公司 Method and device for balancing load as well as handheld terminal and communication equipment
CN102480529A (en) * 2010-11-24 2012-05-30 北京无线恒远科技有限公司 Domain name analysis method and domain name analysis server for realizing load balance of wide area network
CN101631143B (en) * 2009-08-27 2012-07-04 中兴通讯股份有限公司 Multi-server system in load-balancing environment and file transmission method thereof
CN102843401A (en) * 2011-06-23 2012-12-26 中兴通讯股份有限公司 Content distribution method and system
CN103024018A (en) * 2012-12-04 2013-04-03 北京蓝汛通信技术有限责任公司 Method and device for operating multiple content distribution network (CDN) service processes in single device
WO2013097525A1 (en) * 2011-12-28 2013-07-04 华为技术有限公司 Method and system for transmitting network video
CN103312629A (en) * 2012-03-09 2013-09-18 深圳市腾讯计算机系统有限公司 Method, device and system of CDN (content delivery network) flow distribution
CN103441906A (en) * 2013-09-25 2013-12-11 哈尔滨工业大学 System for detecting abnormity of proxy cache cluster based on automatic computing
CN103747043A (en) * 2013-12-24 2014-04-23 乐视网信息技术(北京)股份有限公司 CDN server dispatching method, CDN control center and system
CN103795766A (en) * 2012-11-02 2014-05-14 腾讯科技(深圳)有限公司 P2SP (Peer to Server & Peer) multi-source download system
CN104320455A (en) * 2014-10-23 2015-01-28 京信通信系统(中国)有限公司 Data dispatching method, server and system
CN104735088A (en) * 2015-04-16 2015-06-24 北京金山安全软件有限公司 Server node adjusting method and device in CDN (content delivery network)
CN104780210A (en) * 2015-04-13 2015-07-15 杭州华三通信技术有限公司 Load balancing method and device
CN105207947A (en) * 2015-08-28 2015-12-30 网宿科技股份有限公司 rogressive flow scheduling method and system capable of filtering vibration
CN106034040A (en) * 2015-03-13 2016-10-19 阿里巴巴集团控股有限公司 Scanning node control method, scanning node control device and scanning node control system
CN106453669A (en) * 2016-12-27 2017-02-22 Tcl集团股份有限公司 Load balancing method and server
CN107357871A (en) * 2017-07-04 2017-11-17 东北大学 A kind of successive range query load equalization methods based on feedback towards Storm
CN107567706A (en) * 2015-05-05 2018-01-09 意大利电信股份公司 Session of subscriber in communication network is distributed again
CN107680227A (en) * 2017-10-20 2018-02-09 莲花丝路科技有限公司 A kind of application method and system of smart lock low power consumption remote control system
WO2018086575A1 (en) * 2016-11-10 2018-05-17 深圳创维数字技术有限公司 Method and device for controlling media resource
CN111193800A (en) * 2019-12-31 2020-05-22 厦门汉印电子技术有限公司 Instruction interaction method, device, server and system based on server analysis

Cited By (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101399965B (en) * 2008-09-28 2011-05-11 中兴通讯股份有限公司 Method and device for implementing media content positioning based on EPG server
CN101631143B (en) * 2009-08-27 2012-07-04 中兴通讯股份有限公司 Multi-server system in load-balancing environment and file transmission method thereof
CN102457904A (en) * 2010-10-14 2012-05-16 深圳市泰海科技有限公司 Method and device for balancing load as well as handheld terminal and communication equipment
CN102480529A (en) * 2010-11-24 2012-05-30 北京无线恒远科技有限公司 Domain name analysis method and domain name analysis server for realizing load balance of wide area network
CN102118433A (en) * 2010-12-27 2011-07-06 网宿科技股份有限公司 Multiple-tier distributed cluster system
CN102843401A (en) * 2011-06-23 2012-12-26 中兴通讯股份有限公司 Content distribution method and system
CN102843401B (en) * 2011-06-23 2017-12-26 中兴通讯股份有限公司 A kind of content distribution method and system
CN102404229A (en) * 2011-12-14 2012-04-04 华为技术有限公司 System, device and method for load balancing
WO2013097525A1 (en) * 2011-12-28 2013-07-04 华为技术有限公司 Method and system for transmitting network video
US9693092B2 (en) 2011-12-28 2017-06-27 Huawei Technologies Co., Ltd. Method and system for transmitting network video
CN103312629A (en) * 2012-03-09 2013-09-18 深圳市腾讯计算机系统有限公司 Method, device and system of CDN (content delivery network) flow distribution
CN103795766A (en) * 2012-11-02 2014-05-14 腾讯科技(深圳)有限公司 P2SP (Peer to Server & Peer) multi-source download system
CN103795766B (en) * 2012-11-02 2018-11-23 腾讯科技(深圳)有限公司 A kind of P2SP multi-source download system
CN103024018A (en) * 2012-12-04 2013-04-03 北京蓝汛通信技术有限责任公司 Method and device for operating multiple content distribution network (CDN) service processes in single device
CN103441906A (en) * 2013-09-25 2013-12-11 哈尔滨工业大学 System for detecting abnormity of proxy cache cluster based on automatic computing
CN103441906B (en) * 2013-09-25 2016-08-24 哈尔滨工业大学 Based on from the proxy caching cluster abnormality detection system of host computer
CN103747043A (en) * 2013-12-24 2014-04-23 乐视网信息技术(北京)股份有限公司 CDN server dispatching method, CDN control center and system
CN104320455A (en) * 2014-10-23 2015-01-28 京信通信系统(中国)有限公司 Data dispatching method, server and system
CN104320455B (en) * 2014-10-23 2018-05-01 京信通信系统(中国)有限公司 A kind of data distributing method, server and system
CN106034040A (en) * 2015-03-13 2016-10-19 阿里巴巴集团控股有限公司 Scanning node control method, scanning node control device and scanning node control system
CN106034040B (en) * 2015-03-13 2019-07-02 阿里巴巴集团控股有限公司 The control method of scan node, apparatus and system
CN104780210A (en) * 2015-04-13 2015-07-15 杭州华三通信技术有限公司 Load balancing method and device
CN104735088A (en) * 2015-04-16 2015-06-24 北京金山安全软件有限公司 Server node adjusting method and device in CDN (content delivery network)
CN104735088B (en) * 2015-04-16 2018-09-11 北京金山安全软件有限公司 Server node adjusting method and device in CDN (content delivery network)
CN107567706A (en) * 2015-05-05 2018-01-09 意大利电信股份公司 Session of subscriber in communication network is distributed again
CN105207947A (en) * 2015-08-28 2015-12-30 网宿科技股份有限公司 rogressive flow scheduling method and system capable of filtering vibration
CN105207947B (en) * 2015-08-28 2018-12-04 网宿科技股份有限公司 A kind of gradual traffic scheduling method and system of filtering shake
WO2018086575A1 (en) * 2016-11-10 2018-05-17 深圳创维数字技术有限公司 Method and device for controlling media resource
CN106453669A (en) * 2016-12-27 2017-02-22 Tcl集团股份有限公司 Load balancing method and server
CN106453669B (en) * 2016-12-27 2020-07-31 Tcl科技集团股份有限公司 Load balancing method and server
CN107357871A (en) * 2017-07-04 2017-11-17 东北大学 A kind of successive range query load equalization methods based on feedback towards Storm
CN107357871B (en) * 2017-07-04 2020-08-11 东北大学 Storm-oriented continuous range query load balancing method based on feedback
CN107680227A (en) * 2017-10-20 2018-02-09 莲花丝路科技有限公司 A kind of application method and system of smart lock low power consumption remote control system
CN111193800A (en) * 2019-12-31 2020-05-22 厦门汉印电子技术有限公司 Instruction interaction method, device, server and system based on server analysis

Similar Documents

Publication Publication Date Title
CN101119359A (en) Policy based service load balancing method
CN100484069C (en) File data distributing method and relative device
US8250211B2 (en) Automatic migration of data via a distributed computer network
CN102547478B (en) Triggered slice on-demand system and method of streaming media based on CDN (Content Distribution Network)
US9167049B2 (en) Content distribution network supporting popularity-based caching
JP5068858B2 (en) Distributing content data packages originated by users of a super peer-to-peer network
CN101616170B (en) Method for supplying media stream service and system thereof
CN100463516C (en) An interactive network TV system
CN101729357B (en) Method and device for storage processing and service processing of media files and server cluster
CN106031130A (en) Content delivery network architecture with edge proxy
CN104735044A (en) Streaming media live broadcast method and system
CA2840106A1 (en) Offline download method, multimedia file download method and system thereof
KR20130040236A (en) Method, server and terminal for audio and video on demand
CN101594292A (en) Content delivery method, service redirection method and system, node device
CN101188736A (en) Stream media ordering system and method with STB as the server
CN102395047A (en) Method and equipment for deleting multi-media contents in content publishing network nodes
TW201234194A (en) Data stream management system for accessing mass data
US20160381127A1 (en) Systems and methods for dynamic networked peer-to-peer content distribution
CN104426718A (en) Data monitoring server, cache server and redirection downloading method
CN102761572A (en) Method and system for dynamically adjusting data flow
CN106789956A (en) A kind of P2P order methods and system based on HLS
Wang et al. PLVER: Joint stable allocation and content replication for edge-assisted live video delivery
CN111225248B (en) On-demand content management method and content distribution network on-demand server
CN105284118A (en) Content provision device, content provision method, program, terminal device, and content provision system
CN100596191C (en) Stream media ordering system and method with TV set as the server

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Open date: 20080206