CN104184673A - Load balancing method based on instant communication service and system - Google Patents
Load balancing method based on instant communication service and system Download PDFInfo
- Publication number
- CN104184673A CN104184673A CN201310196308.3A CN201310196308A CN104184673A CN 104184673 A CN104184673 A CN 104184673A CN 201310196308 A CN201310196308 A CN 201310196308A CN 104184673 A CN104184673 A CN 104184673A
- Authority
- CN
- China
- Prior art keywords
- user
- subregion
- server
- service server
- mark
- 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.)
- Granted
Links
Abstract
The invention discloses a load balancing method based on instant communication service and a system, and relates to the instant communication field. The method comprises steps: a social network graph is built according to the instant communication service; zoning is carried out on the social network graph; zones are adjusted to enable the zones and distribution and scale of the service server to be matched; a zone identifier of the user is exchanged to enable the user to fall a server of to-be moved zone service after load balancing hash operation; and dynamic migration is carried out on the user according to the zone identifier of the user. Through carrying out social network zoning on the instant communication relationship among users, users in close relationship are gathered in one zone, the best server load distribution view is thus acquired, the user is migrated to a new home server according to the graph, resource consumption due to cross-zone communication of the server can be reduced, true load balancing of the system at the global level is achieved, and the problem of unbalanced communication among servers due to unreasonable user zoning in the instant communication system can be solved.
Description
Technical field
The present invention relates to instant messaging field, particularly a kind of load-balancing method and system based on instant messaging service.
Background technology
The large-scale instant message system of the Internet level is all supported large-scale consumer by disposing multiple clusters, and corresponding load balancing adopts multistage mode, is generally divided into the overall situation and local two kinds of load sharing policies.Server load balancing algorithm comprises hashing algorithm, precedence algorithm, minimum join algorithm, response speed algorithm, dynamic property allocation algorithm etc.These algorithms are all distributed to the request from user on each server as far as possible uniformly, make the user that in system, each server is received ask treating capacity consistent with the ratio trend of its computing capability.
Instant messaging service belongs to online social networks (Online Social Networks, OSNs) category, the feature of this application is that the request of front end subscriber causes and between back-end server, becomes intensive netted mutually calling, instead of sequence call successively clearly in the past.Therefore, the server of disposing instant messaging service not only needs to process the user's request directly receiving, and also needs inner request of calling mutually between processing server.Cause thus the inner lack of uniformity of calling mutually between server, although cause the number of users distributing on some server node also few, message traffic is very big, has had a strong impact on the balance of the whole network flow.
Existing load balancing mode is considered not enough to the operation characteristic of instant messaging service, need a kind of load balancing scheme of applicable instant messaging service.
Summary of the invention
An embodiment of the present invention technical problem to be solved is: the load balancing scheme that proposes a kind of applicable instant messaging service.
According to the embodiment of the present invention aspect, a kind of load-balancing method based on instant messaging service is proposed, comprising: build social network diagram for instant messaging service; Social network diagram is carried out to subregion; Adjust subregion make the distribution of subregion and service server and scale suitable; The subregion mark of exchanging user makes user carry out falling section post to be moved at service server after the computing of load balancing hash; According to user's subregion mark, user is carried out to dynamic migration.
According to the embodiment of the present invention on the other hand, propose a kind of SiteServer LBS based on instant messaging service, comprising: partitioned server and adjustment of load server; Partitioned server is for building social network diagram for instant messaging service; Social network diagram is carried out to subregion; Adjust subregion make the distribution of subregion and service server and scale suitable; The subregion mark of exchanging user makes user carry out falling section post to be moved at service server after the computing of load balancing hash; Adjustment of load server carries out dynamic migration for the subregion mark according to user to user.
The present invention carries out social activity network partition by the instant communication relation between user, make tight user gather same subregion, thereby obtain optimum server load distribution view, and according to this figure, user is moved on new attribution server, to reduce the resource consumption of server cross-regional communication, make system reach real load balancing in global level, solved the unbalanced problem of communicating by letter between the server causing because user partition is unreasonable in instantaneous communication system.
By the detailed description to exemplary embodiment of the present invention referring to accompanying drawing, it is clear that further feature of the present invention and advantage thereof will become.
Brief description of the drawings
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 is the schematic flow sheet that the present invention is based on an embodiment of load-balancing method of instant messaging service.
Fig. 2 is the load balancing contrast schematic diagram between the present invention and prior art.
Fig. 3 is that the present invention identifies by exchanging subregion the schematic diagram that carries out user's migration.
Fig. 4 is that the present invention is serving the method flow schematic diagram that in not interruption situation, user is carried out dynamic migration.
Fig. 5 is the structural representation that the present invention is based on an embodiment of SiteServer LBS of instant messaging service.
Fig. 6 is the structural representation that the present invention is based on another embodiment of SiteServer LBS of instant messaging service.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiment.Illustrative to the description only actually of at least one exemplary embodiment below, never as any restriction to the present invention and application or use.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.
The large-scale instant message system of the Internet level, user base number is large, number of servers is many, distributional region is wide, if can not reasonably make the user of communications-intensive focus on same region to user partition, the transregional traffic can be very large, causes entire system load imbalance.The present invention proposes a kind of load balancing scheme of applicable instant messaging service, describe in detail below.
Fig. 1 is the schematic flow sheet that the present invention is based on an embodiment of load-balancing method of instant messaging service.
As shown in Figure 1, the load-balancing method of the present embodiment comprises the following steps:
Step 101, builds social network diagram for instant messaging service;
Step 102, carries out subregion to social network diagram;
Step 103, adjust subregion make the distribution of subregion and service server and scale suitable;
Step 104, the subregion mark of exchanging user makes user carry out falling section post to be moved at service server after the computing of load balancing hash;
Step 105, carries out dynamic migration according to user's subregion mark to user.
Above-described embodiment, carry out social activity network partition by the instant communication relation between user, make tight user gather same subregion, thereby obtain optimum server load distribution view, and according to this figure, user is moved on new attribution server, to reduce the resource consumption of server cross-regional communication, make system reach real load balancing in global level, solve the unbalanced problem of communicating by letter between the server causing because user partition is unreasonable in instantaneous communication system.
Fig. 2 is the load balancing contrast schematic diagram between the present invention and prior art.After adopting prior art equilibrium, on server A, B, C, there are respectively 5 users, but due to user A and B in server A, with user C, D and the E in server B, and user F close relation on server C, communication between user is more frequent, causes the resource consumption of server cross-regional communication many.After adopting equilibrium of the present invention, readjust the distribution of user on server according to subregion, user A, B, C, D and E in close relations focus in same server A, thereby have reduced the resource consumption of server cross-regional communication, make system reach real load balancing in global level.
For step 101, the present invention proposes a kind of exemplary method of building social network diagram, comprise: obtain instant messaging service relation between user, relationship type, the frequency of communicating by letter with time correlation from database server, according to the instant messaging service relation between user, relationship type, build various social network diagram with the frequency of communicating by letter of time correlation.Relationship type wherein can be for example good friend's relation, classmate's relation, Peer Relationships etc.In addition, while existing instant messaging service to be related between judgement user, a minimum threshold that has relation can be set, for example can set and have the minimum threshold of relation is 30 message every month, if be less than 30 message, can think and not have instant messaging service relation between two users, otherwise, can think and have instant messaging service relation between two users.
For step 102, how in these extremely huge and variable 1,000,000 of large-scale instant message system online social networks collection of illustrative plates more than node, carrying out various dimensions quick clustering and dynamically adjust user partition, is a very complicated and thorny global question from engineering viewpoint.For this problem, the present invention uses for reference node clustering and the subgraph discovery algorithm in social network analysis (SNA) field, a kind of exemplary social network diagram partition method is proposed, comprise: according to k-core algorithm, social network diagram is carried out to subregion, k-core algorithm refer to any one user in subregion all at least with subregion in other k user's opening relationships, k wherein represents the default number of degrees.It should be noted that, the weight of relation can be set as required, and for example, the good friend who contacted 10 times in month just sets up a relation, although and some user is also good friend but is less than 10 times owing to contacting in one month, opening relationships not between user and these good friends.
For step 103, the present invention proposes a kind of method of exemplary adjustment subregion, comprising: adjust partition size by the value that changes above-mentioned default number of degrees k, make the distribution of subregion and service server and scale suitable.
For step 104, the key(keyword using while carrying out Hash load balancing due to user) be not user ID (or claim user ID) but user's subregion mark (or claiming partition id), this subregion mark is variable with the binding relationship of user ID, therefore the present invention can be by the mapping relations of amendment user ID and subregion mark, and user is fallen on service server to be migrated in the time that the hash of carrying out load balancing is calculated.
Because Hash is irreversible, we cannot know all partition ids that server is corresponding, but can know a part of partition id by the customer group of falling this server, so the present invention proposes following first method, by partition id group known between continuous exchange server, make user be finally allocated to the partition id of destination server.
First method, user's former subregion mark is replaced with to new subregion mark, this new subregion mark can make known the to be moved out subregion mark of section post to be moved in service server, so that user uses subregion mark to carry out falling section post to be moved at service server after the computing of load balancing hash.
For example, as shown in Figure 3, user b wants to move to server B from server A, need to from the partition id to be moved out of B server, choose a partition id of replacing the former server A in user b place, so the partition id of user b changes 4 into, the server B of user b can being moved into.Again for example, user d wants to move to server C from server B, need to from the partition id to be moved out of C server, choose a partition id of replacing the former server B in user d place, so the partition id of user d changes 7 into, server C can move into user d.Subregion after other users exchange identifies as shown in Figure 3, repeats no longer one by one here.
The method of exchanging can not guarantee that all users obtain desirable partition id, and this limitation can be laid in voluntarily Free Partition ID by system and is resolved, i.e. following second method.
Second method, user's former subregion mark is replaced with to new subregion mark, this new subregion mark can make the Free Partition mark of section post to be moved in service server, so that user uses subregion mark to carry out falling section post to be moved at service server after the computing of load balancing hash.
Above-mentioned two kinds of methods, move to new section post after service server user, still can adopt consistency hash algorithm to carry out customer location addressing, do not need the high routing table technology of trace utilization cost, are applicable to very much implementing in engineering.
For step 105, the present invention proposes a kind ofly serving the method for in not interruption situation, user being carried out dynamic migration, as shown in Figure 4, comprises the following steps:
Step 401, adjustment of load server obtains user's user ID, former subregion mark, new subregion mark, carry out Hash operation and obtain the address of the former service server in user place according to former subregion mark, carry out Hash operation and obtain the address of user place new business services device according to new subregion mark;
Step 402, adjustment of load server is according to the address of the former service server in user place, notifies the former service server in user place this user that moves out;
Step 403, adjustment of load server can obtain from database server good friend's (good friend can be tabular form) of user, and the subregion of notifying user good friend place service server to upgrade this user is designated new subregion mark;
Step 404, the former service server in user place can be by connecting in the long of subscription client, notify user to be redirected to user place new business services device, and given load to adjust server be its messaging proxy server, to process its good friend simultaneously also in the situation of migration; The message that forwarding good friend issues user is to adjustment of load server;
Step 405, the new subregion mark of user good friend place service server recording user, is used old subregion to identify to user and sends message, until receive that re-using new subregion after the notice of the new subregion mark of the switching of adjustment of load server identifies to user and send message;
Step 406, the message of issuing the former service server in user place is transmitted to user place new business services device by adjustment of load server, and the message of user place new business services device being issued to this user good friend is transmitted to user good friend place service server;
Step 407, when user good friend place service server all completes after the renewal of new subregion mark, adjustment of load server is to the unified notice of switching new subregion mark that sends of user good friend place service server, after user good friend place service server has notice, use new subregion to identify to user and send message, and notify user place new business services device to initiate a message no longer through adjustment of load transit server.
Above-mentioned moving method, can carry out serving in not interruption situation, and can not miss message all in migration in the situation that user and good friend thereof.
Fig. 5 is the structural representation that the present invention is based on an embodiment of SiteServer LBS of instant messaging service.
As shown in Figure 5, the SiteServer LBS of the present embodiment comprises: partitioned server 501 and adjustment of load server 502; Partitioned server 501 is for building social network diagram for instant messaging service; Social network diagram is carried out to subregion; Adjust subregion make the distribution of subregion and service server and scale suitable; The subregion mark of exchanging user makes user carry out falling section post to be moved at service server after the computing of load balancing hash; Adjustment of load server 502 carries out dynamic migration for the subregion mark according to user to user.
Wherein, partitioned server 501 can comprise that social network diagram builds module, for building social network diagram for instant messaging service, specifically comprise: obtain instant messaging service relation between user, relationship type, the frequency of communicating by letter with time correlation from database server, according to the instant messaging service relation between user, relationship type, build social network diagram with the frequency of communicating by letter of time correlation.
Wherein, partitioned server 501 can comprise division module, for social network diagram is carried out to subregion, specifically comprise: according to k-core algorithm, social network diagram is carried out to subregion, k-core algorithm refer to any one user in subregion all at least with subregion in other k user's opening relationships, k wherein represents the default number of degrees.
Wherein, partitioned server 501 can comprise subregion adjusting module, for adjust subregion make the distribution of subregion and service server and scale suitable, specifically comprise: preset the value of number of degrees k and adjust partition size by changing, make the distribution of subregion and service server and scale suitable.
Wherein, partitioned server 501 can comprise subregion mark adjusting module, make user carry out falling section post to be moved at service server after the computing of load balancing hash for the subregion mark of exchanging user, specifically comprise: user's former subregion mark is replaced with to new subregion mark, this new subregion is designated a Free Partition mark or known the to be moved out subregion mark of section post to be moved in service server, so that user uses subregion mark to carry out falling section post to be moved at service server after the computing of load balancing hash.
Fig. 6 is the structural representation that the present invention is based on another embodiment of SiteServer LBS of instant messaging service.As shown in Figure 6, this system also comprises service server 503.In dynamic migration process:
Adjustment of load server 502, for obtain user's user ID from partitioned server 501, former subregion mark, new subregion mark, carry out Hash operation and obtain the address of the former service server in user place according to former subregion mark, carry out Hash operation and obtain the address of user place new business services device according to new subregion mark, according to the address of the former service server in user place, notify the former service server in user place this user that moves out, obtain user's good friend from database server, the subregion of notifying user good friend place service server to upgrade this user is designated new subregion mark,
The former service server in user place, for notifying user to be redirected to user place new business services device, and given load is adjusted the proxy server that server is its messaging;
User good friend place service server, for the new subregion mark of recording user, uses old subregion to identify to user and sends message, until receive that re-using new subregion after the notice of the new subregion mark of the switching of adjustment of load server identifies to user and send message;
The former service server in user place and user good friend place service server are all designated 503 in Fig. 6, no longer distinguish.
Adjustment of load server 502, for the message of issuing the former service server in user place is transmitted to user place new business services device, the message of user place new business services device being issued to this user good friend is transmitted to user good friend place service server, when user good friend place service server all completes after the renewal of new subregion mark, to the unified notice of switching new subregion mark that sends of user good friend place service server.
In the time of specific implementation, partitioned server, because calculating scale is very large, algorithm often changes, can adopt oracle to realize; Adjustment of load server can be realized based on server framework netty transformation; Service server (as openfire) need to increase restful interface and send notification instruction for adjustment of load server.
Above-mentioned SiteServer LBS, carry out social activity network partition by the instant communication relation between user, make tight user gather same subregion, thereby obtain optimum server load distribution view, and according to this figure, user is moved on new attribution server, to reduce the resource consumption of server cross-regional communication, make system reach real load balancing in global level, solve the unbalanced problem of communicating by letter between the server causing because user partition is unreasonable in instantaneous communication system.And, move to new section post after service server user, still can adopt consistency hash algorithm to carry out customer location addressing, do not need the high routing table technology of trace utilization cost, be applicable to very much implementing in engineering.In addition, user migration can be carried out serving in not interruption situation, and can not miss message all in migration in the situation that user and good friend thereof.
One of ordinary skill in the art will appreciate that all or part of step that realizes above-described embodiment can complete by hardware, also can carry out the hardware that instruction is relevant by program completes, described program can be stored in a kind of computer-readable recording medium, the above-mentioned storage medium of mentioning can be read-only memory, disk or CD etc.
The foregoing is only preferred embodiment of the present invention, in order to limit the present invention, within the spirit and principles in the present invention not all, any amendment of doing, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.
Claims (12)
1. the load-balancing method based on instant messaging service, comprising:
Build social network diagram for instant messaging service;
Social network diagram is carried out to subregion;
Adjust subregion make the distribution of subregion and service server and scale suitable;
The subregion mark of exchanging user makes user carry out falling section post to be moved at service server after the computing of load balancing hash;
According to user's subregion mark, user is carried out to dynamic migration.
2. method according to claim 1, is characterized in that, described for instant messaging service build social network diagram comprise:
According to the instant messaging service relation between user, relationship type, build social network diagram with the frequency of communicating by letter of time correlation.
3. method according to claim 1, is characterized in that, describedly social network diagram is carried out to subregion comprises:
According to k-core algorithm, social network diagram is carried out to subregion, k-core algorithm refer to any one user in subregion all at least with subregion in other k user's opening relationships, k wherein represents the default number of degrees.
4. method according to claim 3, is characterized in that, described adjustment subregion makes the distribution of subregion and service server and scale is suitable comprises:
Adjust partition size by the value that changes default number of degrees k, make the distribution of subregion and service server and scale suitable.
5. method according to claim 1, is characterized in that, described exchange user's subregion mark makes user carry out falling section post to be moved into after the computing of load balancing hash to comprise at service server:
User's former subregion mark is replaced with to new subregion mark, this new subregion is designated a Free Partition mark or known the to be moved out subregion mark of section post to be moved in service server, so that user uses subregion mark to carry out falling section post to be moved at service server after the computing of load balancing hash.
6. method according to claim 1, is characterized in that, the described mark of the subregion according to user is carried out dynamic migration to user and comprised:
Adjustment of load server obtains user's user ID, former subregion mark, new subregion mark, carry out Hash operation and obtain the address of the former service server in user place according to former subregion mark, carry out Hash operation and obtain the address of user place new business services device according to new subregion mark, according to the address of the former service server in user place, notify the former service server in user place this user that moves out, obtain user's good friend, the subregion of notifying user good friend place service server to upgrade this user is designated new subregion mark;
The former service server in user place notifies user to be redirected to user place new business services device, and given load is adjusted the proxy server that server is its messaging;
The new subregion mark of user good friend place service server recording user, is used old subregion to identify to user and sends message, until receive that re-using new subregion after the notice of the new subregion mark of the switching of adjustment of load server identifies to user and send message;
The message of issuing the former service server in user place is transmitted to user place new business services device by adjustment of load server, the message of user place new business services device being issued to this user good friend is transmitted to user good friend place service server, when user good friend place service server all completes after the renewal of new subregion mark, to the unified notice of switching new subregion mark that sends of user good friend place service server.
7. the SiteServer LBS based on instant messaging service, comprising: partitioned server and adjustment of load server;
Partitioned server is for building social network diagram for instant messaging service; Social network diagram is carried out to subregion; Adjust subregion make the distribution of subregion and service server and scale suitable; The subregion mark of exchanging user makes user carry out falling section post to be moved at service server after the computing of load balancing hash;
Adjustment of load server carries out dynamic migration for the subregion mark according to user to user.
8. system according to claim 7, it is characterized in that, described partitioned server comprises that social network diagram builds module, for building social network diagram for instant messaging service, specifically comprise: according to the instant messaging service relation between user, relationship type, with the frequency structure social network diagram of communicating by letter of time correlation.
9. system according to claim 7, it is characterized in that, described partitioned server comprises division module, for social network diagram is carried out to subregion, specifically comprise: according to k-core algorithm, social network diagram is carried out to subregion, k-core algorithm refer to any one user in subregion all at least with subregion in other k user's opening relationships, k wherein represents the default number of degrees.
10. system according to claim 9, it is characterized in that, described partitioned server comprises subregion adjusting module, for adjust subregion make the distribution of subregion and service server and scale suitable, specifically comprise: adjust partition size by the value that changes default number of degrees k, make the distribution of subregion and service server and scale suitable.
11. systems according to claim 7, it is characterized in that, described partitioned server comprises subregion mark adjusting module, make user carry out falling section post to be moved at service server after the computing of load balancing hash for the subregion mark of exchanging user, specifically comprise: user's former subregion mark is replaced with to new subregion mark, this new subregion is designated a Free Partition mark or known the to be moved out subregion mark of section post to be moved in service server, so that user uses subregion mark to carry out falling section post to be moved at service server after the computing of load balancing hash.
12. systems according to claim 7, is characterized in that, described system also comprises service server;
Described adjustment of load server, for obtain user's user ID, former subregion mark, new subregion mark from partitioned server, carry out Hash operation and obtain the address of the former service server in user place according to former subregion mark, carry out Hash operation and obtain the address of user place new business services device according to new subregion mark, according to the address of the former service server in user place, notify the former service server in user place this user that moves out, obtain user's good friend, the subregion of notifying user good friend place service server to upgrade this user is designated new subregion mark;
The former service server in user place, for notifying user to be redirected to user place new business services device, and given load is adjusted the proxy server that server is its messaging;
User good friend place service server, for the new subregion mark of recording user, uses old subregion to identify to user and sends message, until receive that re-using new subregion after the notice of the new subregion mark of the switching of adjustment of load server identifies to user and send message;
Described adjustment of load server, for the message of issuing the former service server in user place is transmitted to user place new business services device, the message of user place new business services device being issued to this user good friend is transmitted to user good friend place service server, when user good friend place service server all completes after the renewal of new subregion mark, to the unified notice of switching new subregion mark that sends of user good friend place service server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310196308.3A CN104184673B (en) | 2013-05-24 | 2013-05-24 | Load-balancing method and system based on instant messaging service |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310196308.3A CN104184673B (en) | 2013-05-24 | 2013-05-24 | Load-balancing method and system based on instant messaging service |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104184673A true CN104184673A (en) | 2014-12-03 |
| CN104184673B CN104184673B (en) | 2017-11-28 |
Family
ID=51965431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310196308.3A Active CN104184673B (en) | 2013-05-24 | 2013-05-24 | Load-balancing method and system based on instant messaging service |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104184673B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107786618A (en) * | 2016-08-31 | 2018-03-09 | 迈普通信技术股份有限公司 | A kind of system of selection for logging in node and device |
| CN108200110A (en) * | 2016-12-08 | 2018-06-22 | 杭州海康威视系统技术有限公司 | A kind of data processing method, apparatus and system |
| CN110830461A (en) * | 2019-10-28 | 2020-02-21 | 杭州涂鸦信息技术有限公司 | Cross-region RPC service calling method and system based on TLS long connection |
| CN110881058A (en) * | 2018-09-06 | 2020-03-13 | 阿里巴巴集团控股有限公司 | Request scheduling method, device, server and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101453370A (en) * | 2007-11-30 | 2009-06-10 | 国际商业机器公司 | Method, equipment and on-line system for user management in on-line system |
| CN102495857A (en) * | 2011-11-21 | 2012-06-13 | 北京新媒传信科技有限公司 | Load balancing method for distributed database |
| CN102577494A (en) * | 2009-09-28 | 2012-07-11 | 瑞典爱立信有限公司 | Method and arrangement for supporting analysis of social networks in a communication network |
-
2013
- 2013-05-24 CN CN201310196308.3A patent/CN104184673B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101453370A (en) * | 2007-11-30 | 2009-06-10 | 国际商业机器公司 | Method, equipment and on-line system for user management in on-line system |
| CN102577494A (en) * | 2009-09-28 | 2012-07-11 | 瑞典爱立信有限公司 | Method and arrangement for supporting analysis of social networks in a communication network |
| CN102495857A (en) * | 2011-11-21 | 2012-06-13 | 北京新媒传信科技有限公司 | Load balancing method for distributed database |
Non-Patent Citations (1)
| Title |
|---|
| 刘磊，倪宏，陈晓: "《对称即时通讯服务器的负载均衡算法研究》", 《计算机工程与应用》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107786618A (en) * | 2016-08-31 | 2018-03-09 | 迈普通信技术股份有限公司 | A kind of system of selection for logging in node and device |
| CN107786618B (en) * | 2016-08-31 | 2020-09-08 | 迈普通信技术股份有限公司 | Method and device for selecting login node |
| CN108200110A (en) * | 2016-12-08 | 2018-06-22 | 杭州海康威视系统技术有限公司 | A kind of data processing method, apparatus and system |
| CN110881058A (en) * | 2018-09-06 | 2020-03-13 | 阿里巴巴集团控股有限公司 | Request scheduling method, device, server and storage medium |
| CN110881058B (en) * | 2018-09-06 | 2022-04-12 | 阿里巴巴集团控股有限公司 | Request scheduling method, device, server and storage medium |
| CN110830461A (en) * | 2019-10-28 | 2020-02-21 | 杭州涂鸦信息技术有限公司 | Cross-region RPC service calling method and system based on TLS long connection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104184673B (en) | 2017-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Aral et al. | A decentralized replica placement algorithm for edge computing | |
| Din et al. | Caching in information-centric networking: Strategies, challenges, and future research directions | |
| US10523748B2 (en) | Managing health status of network devices in a distributed global server load balancing system | |
| EP3087722B1 (en) | Peer-to-peer network prioritizing propagation of objects through the network | |
| CN103825834A (en) | Method and network exchange device for adjusting flow load sharing in EVI | |
| CN104184673A (en) | Load balancing method based on instant communication service and system | |
| CN101808119A (en) | Method and equipment for multiple storage array load balancing | |
| CN105068755A (en) | Data duplicate storage method facing cloud computing content distribution network | |
| Sourlas et al. | Mobility support through caching in content-based publish/subscribe networks | |
| Li et al. | Deployment of edge servers in 5G cellular networks | |
| CN105681413A (en) | Method and device for cooperative processing of data between CDN (Content Delivery Network) and ISP (Internet Service Provider) | |
| CN103001892A (en) | Network resource distribution method and system based on cloud computing | |
| CN106686060A (en) | Content spread method and system | |
| CN110198267A (en) | A kind of traffic scheduling method, system and server | |
| CN1330124C (en) | Method and apparatus for virtualizing network resources | |
| JP5364070B2 (en) | Virtual server management device | |
| Wittenburg et al. | A survey of current directions in service placement in mobile ad-hoc networks | |
| CN105007234A (en) | Load balancing method for global ip scheduling | |
| Turukmane et al. | An Effective Routing Algorithm for Load balancing in Unstructured Peer-to-Peer Networks | |
| Shen et al. | Hash-based proximity clustering for load balancing in heterogeneous DHT networks | |
| Fekih et al. | SDN-based replication management framework for CCN networks | |
| CN113301085B (en) | Scheduling method, device, equipment and storage medium | |
| US10623260B2 (en) | Software defined network (SDN) information distribution across an SDN data-plane | |
| Li et al. | A dynamic load balancing algorithm based on consistent hash | |
| Jiang | Wireless resource management mechanism with green communication for multimedia streaming |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |