CN104158911A - Server cluster-oriented control method and system - Google Patents
Server cluster-oriented control method and system Download PDFInfo
- Publication number
- CN104158911A CN104158911A CN201410439567.9A CN201410439567A CN104158911A CN 104158911 A CN104158911 A CN 104158911A CN 201410439567 A CN201410439567 A CN 201410439567A CN 104158911 A CN104158911 A CN 104158911A
- Authority
- CN
- China
- Prior art keywords
- server
- service
- cluster
- effective
- server cluster
- 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
Links
Landscapes
- Computer And Data Communications (AREA)
Abstract
The invention discloses a server cluster-oriented control method and system. The method comprises the following steps: acquiring update information about a server cluster, wherein the update information is the quantity of valid servers needing to be arranged in the server cluster; updating a pre-stored service list according to the update information about the server cluster, wherein the service list comprises the quantity of service command requests currently started by each server and the valid zone bit of each server in the server cluster, wherein the valid zone bits are used for judging whether the servers are valid or not; transmitting current service command requests to be processed to the valid servers in the service list for processing; and updating the service list according to a processing result. According to the embodiment, the servers in the server cluster are taken as independent nodes, so that the servers can be extended or reduced without restarting the whole server cluster, and the reliability of a system is improved.
Description
Technical field
The present invention relates to communication technical field, relate in particular to a kind of control method and system of service-oriented device cluster.
Background technology
Fast development along with computer technology and the network communications technology, server cluster is widely used, adopt the system of server cluster normally in order to improve the stability of system and the data-handling capacity of network center and service ability, in server cluster, there are a plurality of parallel servers, these servers are worked together and are moved a series of common programs and for example process service command request, and for user, corresponding application provides man-to-man system map with service command request.
In the prior art, server in cluster connects by cable physically, by a special load-balanced server according to the logical relation between server, for every station server assigns access request in whole group system, load-balanced server receives the access request that user sends, access request can be issued one by one to every station server processes, like this, the access request that every station server in server cluster distributes is just comparatively average, example, in server cluster, there are four station servers, be respectively server 1, server 2, server 3, server 4, load-balanced server can be distributed to one by one four station servers by logical order by 100 access request and process, every like this station server is just got 25 access request, the electricity maintenance under temporary needs if certain station server in cluster server breaks down, or because the excessive needs of visit capacity increase the number of servers in cluster temporarily, load-balanced server needs descending business original on electric server issue other servers or new access request is dynamically switched to the server newly adding and processed.
But in the process of present inventor's invention technical scheme in realizing the embodiment of the present application, find that above-mentioned technology at least exists following technical problem:
Increase, reducing server can affect the configuration of whole server cluster, and need to restart whole server cluster, if having new server wants to add in server cluster system, when connecting with load-balanced server, need to revise the configuration of whole group system, need to set up logical relation with existing server, example, in simple terms, if there is new server need to add in server cluster system, can not only power on, otherwise load-balanced server does not judge that new server is whether in the logic of server cluster and can not be new server-assignment visit capacity, therefore server cluster need to be restarted, by new Servers installed, be that server 5 could continue normal work, in like manner, when need to reduce server, also need to change the logical relation between existing server, need to restart whole server cluster equally, then original load on the server of reduction being distributed to other servers processes, in the process that server cluster is restarted, need to spend the regular hour, and need to change the configuration of server cluster system, because the configuration in server cluster renewal process is comparatively complicated, objectively increase the possibility that server cluster system is made mistakes, also reduced reliability and the performance of server system.
Summary of the invention
The embodiment of the present application is by providing a kind of control method and system of service-oriented device cluster, solved and in prior art, increased, reduced the configuration that server affects whole server cluster, need to restart whole group system, the possibility that increase server cluster system is made mistakes, the performance that has also reduced server system, realized can be in guaranteeing server cluster all the other servers unaffected, when not needing to restart server cluster, can upgrade easily and revise number of servers in server cluster.
To achieve these goals, the embodiment of the present application provides a kind of control method of service-oriented device cluster on the one hand, said method comprising the steps of: obtain the lastest imformation of described server cluster, described lastest imformation is in described server cluster, to need to arrange the quantity of effective server; According to the lastest imformation of described server cluster, upgrade the service list prestoring, described service list comprises the service command number of request of the current startup of each server and the effective marker position of each server in described server cluster, and described effective marker position is used for judging that whether described server is effective; Current pending service command request is sent to the effective server process in described service list; According to result, upgrade described service list.
On the other hand, the embodiment of the present application also provides a kind of control system of service-oriented device cluster, and described system comprises as follows:
The first acquisition module, for obtaining the lastest imformation of described server cluster; The first update module, the described lastest imformation of obtaining according to the first acquisition module is upgraded and is pre-stored in the service list in common data library module, and described lastest imformation is in described server cluster, to need to arrange the quantity of effective server; Common data library module, prestore described service list, described service list comprises the service command number of request of the current startup of each server and the effective marker position of each server in described server cluster, and described effective marker position is used for judging that whether described server is effective; Processing module, for current pending service command request being sent to effective server process of described service list, and sends to the second update module result; The second update module, is connected with common data module and processing module, according to the result of processing module feedback, upgrades and is stored in the service list in common data library module.
The technical scheme providing in the embodiment of the present application, at least has following technique effect or advantage:
By obtaining the lastest imformation of described server cluster, described lastest imformation is in described server cluster, to need to arrange the quantity of effective server, can know whether server cluster has change to the quantity of effective server, then according to the lastest imformation of described server cluster, upgrade the service list prestoring, because service list comprises the service command petition of the current startup of each server and the effective marker position of each server in described server cluster, send to the effective server in the service list after renewal to process current pending service command request, then according to result, upgrade described service list, compared to existing technology, every station server in server cluster is independently each other, effective number of servers in server cluster is only subject to the renewal data influence in service list, when the effective number of servers in server cluster has renewal, do not need to restart whole aggregated server system, only need in service list, upgrade data, existing server operation business is not affected, like this, each station server is all an independently node, there is no each other direct logical relation, thereby the server in assurance server cluster is separate, when the quantity of effective server is had to renewal, can not need to restart whole server cluster, do not need to change the configuration of whole system yet, and then reduce the possibility of makeing mistakes of whole group system, increase the system reliability of server cluster.
Accompanying drawing explanation
Fig. 1 is the flow chart of the control method of service-oriented device cluster in the embodiment of the present application;
Fig. 2 is a kind of method flow diagram that generates lastest imformation in the embodiment of the present application;
Fig. 3 is another method flow diagram that generates lastest imformation in the embodiment of the present application;
Fig. 4 is the method flow diagram of searching and determine effective server in the embodiment of the present application;
Fig. 5 is the module diagram of the control system of service-oriented device cluster in the embodiment of the present application;
Fig. 6 is the module diagram of processing module in the embodiment of the present application;
Fig. 7 is a kind of module diagram that generates lastest imformation in the embodiment of the present application;
Fig. 8 is another module diagram that generates lastest imformation in the embodiment of the present application;
?
Embodiment
Technical scheme in the embodiment of the present application is for solving the problems of the technologies described above, and general thought is as follows:
The application is to provide a kind of control method and system of service-oriented device cluster, in server cluster, prestore service list, whether effectively service list comprises the service command number of request of the current startup of each server in server cluster and judges server effective marker position, while having lastest imformation in server cluster, lastest imformation is mainly the quantity of effective server in server cluster, the service list that renewal prestores, server in service list is independently each other in logic, in service list, arbitrary effective server can be as the processing server of processing next service command request, like this, when the effective number of servers in server cluster is had to renewal, do not need to restart whole server cluster, only need in service list, upgrade data, effective server that pending service command request meeting is issued in the service list after renewal is realized alignment processing, thereby the server in assurance server cluster is separate, increase the system reliability of server cluster.
In order to make the technical staff in the technical field of the invention more clearly understand the present invention, below in conjunction with accompanying drawing, by specific embodiment, the technical scheme in the embodiment of the present application is described in detail.
As shown in Figure 1, the invention provides a kind of control method of service-oriented device cluster, comprise the following steps:
S1: obtain the lastest imformation of described server cluster, described lastest imformation is to need to arrange the quantity of effective server in described server cluster;
In this step, effectively server refers to server that can the request of normal process service command, the size of the ability to work that effectively quantity of server can response service device cluster.
Before server cluster is started working, first first obtain the lastest imformation of server cluster, lastest imformation comprises the quantity of effective server that server cluster should arrange, example, if have three station servers in current server cluster in effective operating state, be respectively server 1, server 2, server 3, according to the service command number of request of the actual startup of server cluster, need to adjust the quantity of effective server, if need to increase the work that an effective server carrys out assist server cluster, the quantity that lastest imformation is effective server is 4, if need to reduce the quantity that an effective server reduces effective server of server cluster, the quantity that lastest imformation is effective server is 2.
As a kind of implementation preferably, as Fig. 2, the quantity of the effective server in lastest imformation can determine as follows,
Q1: the present load saturation of obtaining described server cluster;
Q2: judge whether the current load saturation degree of described server cluster is greater than the first predetermined threshold value;
Q3: if so, calculate the number of servers that described server cluster need to increase and generate lastest imformation according to the difference of described load saturation degree and the first predetermined threshold value.
The unit interval of supposing three effective servers in current server cluster under ideal conditions need service command request total amount to be processed be 20, every station server unit interval disposal ability is 5, the disposal ability of three station servers is 15, three station servers far can not meet the service command request to be processed of unit interval domestic demand, now need to be greater than the first default predetermined threshold value according to the load saturation degree of service command request total amount and known three the effective servers of disposal ability, according to the difference of load saturation degree and the first predetermined threshold value is known, need to increase by a station server, be in server cluster, to need to have 4 station servers to process the total amount that just can complete predetermined service command request, if occur that in computational process the number of servers that the difference of load saturation degree and the first predetermined threshold value calculates needs and increases is not the situation of integer, not enough part is calculated by a station server.
As another implementation preferably, as Fig. 3, the quantity of the effective server in lastest imformation also can be determined as follows:
P1: the present load saturation of obtaining described server cluster;
P2: judge whether described load saturation degree is less than the second predetermined threshold value;
P3: if so, calculate and need the number of servers of described server cluster minimizing and generate lastest imformation according to the difference of described load saturation degree and the second predetermined threshold value.
The unit interval of supposing three effective servers in current server cluster under ideal conditions need service command request total amount to be processed be 20, every station server unit interval disposal ability is 10, the disposal ability of three station servers is 30, the processing energy of three station servers will be much larger than the service command request to be processed of unit interval domestic demand, now need to be less than the second predetermined threshold value according to the load saturation degree of service command request total amount and known three the effective servers of disposal ability, according to known minimizing one station server of the difference of load saturation degree and the first predetermined threshold value, also can realize the processing to service command request, be only need to have 2 station servers to process the total amount that just can complete predetermined service command request in server cluster, if occur that in computational process it is not the situation of integer that the difference of load saturation degree and the second predetermined threshold value calculates the number of servers that needs reduce, not enough part can need to be calculated by a station server according to actual conditions.
Certainly, the quantity of effective server in lastest imformation, not only can calculate according to the difference of the current load saturation degree of server cluster and the first predetermined threshold value or the second predetermined threshold value, also can freely arrange according to user's actual demand, by user, estimate and judge the quantity that whether needs increase or reduce effective server in server cluster, and the form with lastest imformation embodies this demand.
S2: upgrade the service list prestoring according to the lastest imformation of described server cluster, described service list comprises the service command petition of current startup and the effective marker position of each server of each server in described server cluster, and described effective marker position is used for judging that whether described server is effective;
As shown in the table, in the public database of server cluster, prestore and record the service command number of request of the current startup of every station server and the service list of effective marker position in current server cluster, certainly, except public database, can store-service list in the storage area of arbitrary station server, do not need load-balanced server of extra configuration, can reduce the cost of server cluster.
Table 1 has recorded the loading condition (load is the service command request of each startup of server) on current Servers-all, and " server effective marker " is that this server of 1 expression is the server of a normal job, 0 represents that this server is an invalid server, known according to table 1 Query Result, server 1,2,3 are effective server, server 1 has started 4 service command requests, and server 2 and server 3 have all started 2 service command requests.
Table 1
| Server sequence number | The service command number of request of current startup | Server effective marker |
| 1 | 4 | 1 |
| 2 | 2 | 1 |
| 3 | 2 | 1 |
When lastest imformation shows that server cluster need to increase by a station server, according to lastest imformation, can upgrade service list, as following table 2:
Table 2
| Server sequence number | The service command number of request of current startup | Server effective marker |
| 1 | 4 | 1 |
| 2 | 2 | 1 |
| 3 | 2 | 1 |
| 4 | 0 | 1 |
Wherein, 4 is the server sequence number newly increasing, because this server 4 has just added, thus on server 4, be not activated any service command request, so the service command number of request of current startup is 0.
When if lastest imformation shows to need to reduce by a station server in server cluster, according to lastest imformation, can upgrade service list, if there are 3 station servers in current server cluster, numbering is respectively 1,2,3, want now, the server unloading that is numbered 2, according to this lastest imformation, can obtain as following table 3:
Table 3
| Server sequence number | The service command number of request of current startup | Server effective marker |
| 1 | 4 | 1 |
| 2 | 2 | 0 |
| 3 | 2 | 1 |
Server is numbered to 2 " server effective marker " and is set to 0, represent that server 2 is invalid servers, now server 2 no longer responds any service command request.
Certainly, also can directly server 2 be deleted from form, the server sequence number of original server 3 is changed to server 2; This specifically will determine according to actual conditions, iff not wanting to allow server 2 work, changes server effective marker position.
S3: current pending service command request is sent to the effective server process in described service list;
System judges the service command request to be processed of current need, need to be to the corresponding processing server of service command request arrangement, relatively and prior art, do not need special load-balanced server receive user's service command request and be responsible for distribution, only need a receiver module with the request of reception service command to process request.
When current pending service command request is carried out to allocation process, current pending service command request can be issued to an arbitrary effective server process in server cluster.
Preferably, as Fig. 4, can to pending service command, distribute as follows,
M1: obtain current pending service command request;
M2: search in described service list and start the minimum and effective server of service command number of request as the processing server of processing described service command request;
M3: send to described processing server to process the described service command request receiving;
According to load balancing principle, when the server cluster after upgrading is distributed to new service command request, for service command request can be carried out to the distribution of balance, therefore basic principle is that service command request is shared on a plurality of servers and carried out, therefore new service command request is given to the less effective server of load, can improve the processing speed of server cluster, raising efficiency, can make the maximized utilization of server cluster resource.
S4: upgrade described service list according to result.
Example, if lastest imformation neither needs to increase effective server, do not need to reduce the quantity of effective server yet, so as can be seen from Table 1, the minimum server of current service command request number is server 2 and server 3, the service command number of request that load capacity starts is all 2, therefore new service command request can be issued any one in two-server, in general, can be according to the front and back order of the current sequence number of server, issue server 2 and process, then updating form 1, as follows
Table 1 after renewal
| Server sequence number | Current service command number of request | Server effective marker |
| 1 | 4 | 1 |
| 2 | 3 | 1 |
| 3 | 2 | 1 |
Example, as shown in table 2, when needs increase server in server cluster, user's service command request sends to all servers, due to the server of server 4 for newly adding, be not activated service command request, that load is the minimum server of service command number of request, according to the load balancing principle to server cluster (the server response access request that load is minimum), this request is processed by server 4, and other server is not processed, and server 4 is handled after this service command request, meeting updating form 2, the table 2 after renewal is as follows:
Table 2 after renewal
| Server sequence number | The service command number of request of current startup | Server effective marker |
| 1 | 4 | 1 |
| 2 | 2 | 1 |
| 3 | 2 | 1 |
| 4 | 1 | 1 |
Now, due to server 4 treated a service command request, therefore the current service command number of request of server becomes 1,
In server cluster, dynamically increase a server, only need to revise the data that record in a service list, do not need to restart other server in server cluster, if and new server process is all new service command request, thus on other server current service functions and performance without any impact.
Example, as shown in table 3, when needs reduce server in server cluster, by table 3, learn, server 2 is current is being that 2 service command requests are served, when needs server 2 quits work, the service command request meeting now having started on server 2 resends to other servers, because server 2 is invalid server now, so server 2 is response request no longer, this request is by server 1 or server 3 responses, because server 3 is 2 with the service command request quantity starting, so respond the service command request on former server 2 by server 3.On server 2,2 service command requests of operation have been taken over regard to serviced device 3, and server 2 just can be realized safe unloading, according to this lastest imformation updating form 3, as follows:
Table 3 after renewal
| Server sequence number | The service command number of request of current startup | Server effective marker |
| 1 | 4 | 1 |
| 2 | 0 | 0 |
| 3 | 4 | 1 |
The server flow process reducing in server cluster is simple, only need to revise the data that record in a service list, and the request on former reduction server is issued to other server process again, then needs the server of reduction just can unload safely.
It should be noted that, the effective server difference increasing in server cluster with expansion is, effective server in reduction server cluster need to be restarted the service command request having started on this server, after service command request is restarted, the service command request moving on this server will be replaced by an other station server, thereby this server is lower electricity smoothly, unloading, and service command request is restarted very fast, just complete several seconds, service command requested service is not had to impact substantially, and the time that server cluster of the prior art is restarted is slower, the time of restarting much larger than service command request.
After handling new service command request, need to service list, upgrade according to result, after upper once service command request receiving, can on the existing service list basis of having upgraded, upgrade again the circulation in implementation method meaning like this.
Preferably, after the step of the described renewal of the lastest imformation according to described server cluster service list, each server in described server cluster can also be arranged in order according to described current service command request number order from more to less.
In the server cluster system of practical application, such as net cast business, sending short messages in groups business, when railway booking business etc. need to be to mass data fast processing, number of servers in whole server cluster system is very many, the listings data amount prestoring is also larger, when server cluster system has more new data, service list is arranged by current service order load capacity order from more to less, can greatly accelerate the relatively how many speed of load capacity, find rapidly the minimum server of service command load capacity, simultaneously, because the minimum server of service command load capacity may more than one, like this, when number of servers in service list is more, also can dwindle data search scope, accelerate the processing speed of server cluster.
As preferred embodiment, searching the minimum and effective server of service command number of request in service list specifically comprises as the step of processing the processing server of described service command request: to starting the minimum effective Servers installed priority of service command number of request in service list, get effective server of described priority maximum as described processing server.
As shown in table 4
Table 4
| Server sequence number | The service command number of request of current startup | Server effective marker | Priority |
| 1 | 5 | 1 | ? |
| 2 | 2 | 1 | ? |
| 3 | 1 | 1 | A |
| 4 | 1 | 1 | B |
| 5 | 1 | 1 | C |
| 6 | 1 | 1 | D |
| 7 | 1 | 1 | E |
Known according to table 4, due to server 3, 4, 5, 6, the service command number of request of 7 current startup is 1, because the load capacity of this 5 station server is that service command number of request equates, be the minimum server of load capacity, therefore in time application, all design different priority can to every station server, as ABCDE, can stipulate according to actual needs the priority orders of ABCDE, as A represents that priority is the highest, therefore new pending service command request can be issued to the server 3 that priority A is corresponding processes, being set in this and not doing too much restriction of priority.
In addition, also have a kind of preferred mode, pending service command request distributed, can from a certain station server, new service command request all be issued to this station server, to this station server, can do the judgement of following steps:
If the load saturation degree of first server is greater than the predetermined upper limit, the one or more service command requests that start in described first server are issued to other the effective server process in described server cluster; Wherein, described first server is arbitrary effective server in described server cluster.
Like this, the service command number of request starting when a certain station server surpasses upper prescribing a time limit, server cluster need to be dispatched the load of this station server, process for arbitrary that the service command request of this station server is issued in remaining effective server, because in actual use, in server cluster, the actual treatment ability of every station server is not completely equal, example, the disposal ability of 1 unit interval of server is 10, the disposal ability of 2 unit interval of server is 50, now, if the service command request having started on server 1 is 12, the service command request starting on server 2 is 30, although the least-loaded of server 1, if but have new service command request to need to process, now the load saturation degree of server 1 has been greater than the predetermined upper limit, it is actual disposal ability, issuing server 1 processes just improper again, now, new service command request need to be issued to the server 2 that disposal ability is stronger, thereby avoid causing service command request on server to block up, load saturation degree between can balance server, thereby can be processed as early as possible new service command request, improved the treatment effeciency of server cluster.
Based on above embodiment, by obtaining the lastest imformation of described server cluster, can know whether server cluster has renewal to the quantity of effective server, according to the lastest imformation of server cluster, upgrade the service list prestoring, send to the effective server in the service list after renewal to process current pending service command request, then according to result, upgrade service list, server in server cluster is independently each other, Servers-all is subject to the renewal data influence in service list, when upgrading for example expansion or reduction server, do not need to restart whole aggregated server system, only need in service list, upgrade data, if need to newly add server, new server powers on and just can work, existing server operation business is not affected, if reduction server does not need to restart whole group system yet, only need in service list, upgrade data, the business of moving on this server will be replaced by an other station server, thereby this server is lower electricity smoothly, unloading, each station server is all an independently node, there is no each other direct logical relation, thereby the server in assurance server cluster is separate, in expansion, in the time of reduction server, reduce the possibility of makeing mistakes of whole group system, increase the system reliability of server cluster.
Based on same inventive concept, the present embodiment also provides a kind of system of dynamic load leveling of service-oriented device cluster to comprise: the first acquisition module, for obtaining the lastest imformation of server cluster; The first update module, the lastest imformation of obtaining according to the first acquisition module is upgraded and is pre-stored in the service list in common data library module, and lastest imformation is in server cluster, to need to arrange the quantity of effective server; Common data library module, prestores service list, and described service list comprises the service command number of request of the current startup of each server and the effective marker position of each server in server cluster, and effective marker position is used for judging that whether server is effective; Processing module, for current pending service command request being sent to effective server process of service list, and sends to the second update module result; The second update module, is connected with common data module and processing module, according to the result of processing module feedback, upgrades and is stored in the service list in common data library module.The connected mode of native system is shown in Fig. 5, the first acquisition module is connected with the first update module, the first acquisition module is issued the first update module lastest imformation, the first update module is connected with public database, the list of common data library module store-service, the first update module is upgraded service list according to lastest imformation, and processing module is connected with common data library module, according to service list, pending service command request is issued to effective server process, and result is issued to the second update module; The second update module is connected with common data library module and processing module, and the result that receiving processing module sends is upgraded service list.
Preferably, as shown in Figure 6, processing module inside also comprises as lower module: the second acquisition module, for obtaining current pending service command request, enquiry module is for transferring the service list that common data library module is stored, and judge that the minimum effective server of the service command number of request that starts in service list is as processing server, sending module is connected with the second acquisition module, and the current pending service command request that the second acquisition module is got is issued described processing server and processed.
As preferred embodiment a kind of, as shown in Figure 7, native system also comprises: the 3rd acquisition module, for obtaining the present load saturation of server cluster; The first judge module, for judging whether load saturation degree is greater than the first predetermined threshold value; The first generation module, is connected with the first judge module, if load saturation degree is greater than the first predetermined threshold value, according to the difference of load saturation degree and the first predetermined threshold value, calculates the number of servers that described server cluster need to increase and generates lastest imformation.
The 3rd acquisition module is connected with common data library module, according to service list, obtain the present load saturation of server cluster, the first judge module is connected with the 3rd acquisition module, according to the quantity of the effective server that needs to increase in the difference calculation server cluster of the first predetermined threshold value prestoring in the first judge module and the present load saturation obtained from the 3rd acquisition module and generate lastest imformation and send to the first acquisition module.
As another preferred embodiment, as shown in Figure 8, native system also comprises: the 4th acquisition module, for obtaining the present load saturation of server cluster; The second judge module, for judging whether load saturation degree is less than the second predetermined threshold value; The second generation module, is connected with the second judge module, if load saturation degree is less than the second predetermined threshold value, according to the difference of load saturation degree and the second predetermined threshold value, calculates and needs the number of servers of server cluster minimizing and generate lastest imformation.
The 4th acquisition module is connected with common data library module, according to service list, obtain the present load saturation of server cluster, the second judge module is connected with the 4th acquisition module, according to the quantity of the effective server that needs in the difference calculation server cluster of the second predetermined threshold value prestoring in the present load saturation of obtaining from the 4th acquisition module and the second judge module to reduce and generate lastest imformation and send to the first acquisition module.
Preferably, in the processing module of native system, also comprise: priority block, for service list being started to the minimum effective Servers installed priority of service command number of request and module being set, for effective Servers installed of judging described priority maximum as described processing server.
As another implementation preferably, native system also comprises: forwarding module, if for judging that the load saturation degree of first server is greater than the predetermined upper limit, the one or more service command requests that start in described first server are issued to other the effective server process in described server cluster; Wherein, described first server is arbitrary effective server in described server cluster.
In actual use, in server cluster, the actual treatment ability of every station server is not completely equal, the service command number of request starting when a certain station server surpasses upper prescribing a time limit, server cluster need to be dispatched the load of this station server, process for arbitrary that the service command request of this station server is issued in remaining effective server, thereby avoid causing service command request on server to block up, load saturation degree between can balance server, thereby can be processed as early as possible new service command request, improved the treatment effeciency of server cluster.
Based on above embodiment, native system obtains the lastest imformation of described server cluster by the first acquisition module, and send to update module to upgrade lastest imformation to be pre-stored in the service list in common data library module, the service command request that processing module is processed need sends to the effective server in service list to process, then result is issued to the second update module, second upgrades service list, because the server in server cluster works alone each other, Servers-all is subject to the renewal data influence in service list, when the effective number of servers in server cluster is upgraded, for example increase or reduce the quantity of effective server, do not need to restart whole aggregated server system, only need in service list, upgrade data, each station server is all an independently node, the logical relation not having a direct impact each other, thereby guarantee in expansion, in the time of reduction server, reduce the possibility of makeing mistakes of whole group system, increase the system reliability of server cluster.
Those skilled in the art should understand, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt complete hardware implementation example, implement software example or in conjunction with the form of the embodiment of software and hardware aspect completely.And the present invention can adopt the form that wherein includes the upper computer program of implementing of computer-usable storage medium (including but not limited to magnetic disc store, CD-ROM, optical memory etc.) of computer usable program code one or more.
The present invention is with reference to describing according to flow chart and/or the block diagram of the method for the embodiment of the present invention, equipment (system) and computer program.Should understand can be in computer program instructions realization flow figure and/or block diagram each flow process and/or the flow process in square frame and flow chart and/or block diagram and/or the combination of square frame.Can provide these computer program instructions to the processor of all-purpose computer, special-purpose computer, Embedded Processor or other programmable data processing device to produce a machine, the instruction of carrying out by the processor of computer or other programmable data processing device is produced for realizing the device in the function of flow process of flow chart or a plurality of flow process and/or square frame of block diagram or a plurality of square frame appointments.
These computer program instructions also can be stored in energy vectoring computer or the computer-readable memory of other programmable data processing device with ad hoc fashion work, the instruction that makes to be stored in this computer-readable memory produces the manufacture that comprises command device, and this command device is realized the function of appointment in flow process of flow chart or a plurality of flow process and/or square frame of block diagram or a plurality of square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device, make to carry out sequence of operations step to produce computer implemented processing on computer or other programmable devices, thereby the instruction of carrying out is provided for realizing the step of the function of appointment in flow process of flow chart or a plurality of flow process and/or square frame of block diagram or a plurality of square frame on computer or other programmable devices.
Although described the preferred embodiments of the present invention, once those skilled in the art obtain the basic creative concept of cicada, can make other change and modification to these embodiment.So claims are intended to all changes and the modification that are interpreted as comprising preferred embodiment and fall into the scope of the invention.
Obviously, those skilled in the art can carry out various changes and modification and not depart from this present invention
Bright spirit and scope.Like this, if within of the present invention these are revised and modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention is also intended to comprise these changes and modification interior.
Claims (12)
1. a control method for service-oriented device cluster, is characterized in that, said method comprising the steps of:
Obtain the lastest imformation of described server cluster, described lastest imformation is in described server cluster, to need to arrange the quantity of effective server;
According to the lastest imformation of described server cluster, upgrade the service list prestoring, described service list comprises the service command number of request of the current startup of each server and the effective marker position of each server in described server cluster, and described effective marker position is used for judging that whether described server is effective;
Current pending service command request is sent to the effective server process in described service list;
According to result, upgrade described service list.
2. the control method of service-oriented device cluster as claimed in claim 1, is characterized in that, describedly sends in described service list the effectively step of server process to be specially current pending server command request:
Obtain current pending service command request;
Search in described service list and start the minimum and effective server of service command number of request as the processing server of processing described service command request;
Send to described processing server to process the described service command request receiving.
3. the method for the dynamic load leveling of a kind of service-oriented device cluster as claimed in claim 1, it is characterized in that, in the described lastest imformation of obtaining described server cluster, described lastest imformation is also to comprise the steps: before needing to arrange the step of quantity of effective server in described server cluster
Obtain the present load saturation of described server cluster, described load saturation degree is the ratio of the service command number of request of the current startup of described server cluster and the service command request disposal ability of described server cluster unit interval;
Judge whether described load saturation degree is greater than the first predetermined threshold value;
If so, according to the difference of described load saturation degree and the first predetermined threshold value, calculate the number of servers that described server cluster need to increase and generate lastest imformation.
4. the control method of service-oriented device cluster as claimed in claim 1, it is characterized in that, in the described lastest imformation of obtaining described server cluster, described lastest imformation is also to comprise the steps: before needing to arrange the step of quantity of effective server in described server cluster
Obtain the present load saturation of described server cluster;
Judge whether described load saturation degree is less than the second predetermined threshold value;
If so, according to the difference of described load saturation degree and the second predetermined threshold value, calculate and need the number of servers of described server cluster minimizing and generate lastest imformation.
5. the control method of service-oriented device cluster as claimed in claim 1, is characterized in that, described method also comprises:
If the load saturation degree of first server is greater than the predetermined upper limit, the one or more service command requests that start in described first server are issued to other the effective server process in described server cluster; Wherein, described first server is arbitrary effective server in described server cluster.
6. the control method of service-oriented device cluster as claimed in claim 2, it is characterized in that, described in search and in described service list, start the minimum and effective server of service command number of request and specifically comprise as the step of processing the processing server of described service command request:
To starting the minimum effective Servers installed priority of service command number of request in described service list;
Get effective server of described priority maximum as described processing server.
7. a control system for service-oriented device cluster, is characterized in that, described system comprises as follows:
The first acquisition module, for obtaining the lastest imformation of described server cluster;
The first update module, the described lastest imformation of obtaining according to the first acquisition module is upgraded and is pre-stored in the service list in common data library module, and described lastest imformation is in described server cluster, to need to arrange the quantity of effective server;
Common data library module, prestore described service list, described service list comprises the service command number of request of the current startup of each server and the effective marker position of each server in described server cluster, and described effective marker position is used for judging that whether described server is effective;
Processing module, for current pending service command request being sent to effective server process of described service list, and sends to the second update module result;
The second update module, is connected with common data module and processing module, according to the result of described processing module feedback, upgrades and is stored in the service list in described common data library module.
8. the control system of service-oriented device cluster as claimed in claim 7, is characterized in that, described processing module also comprises:
The second acquisition module, for obtaining current pending service command request;
Enquiry module, for searching the minimum and effective server of service command number of request that described service list starts as the processing server of processing described service command request;
Sending module, issues described processing server for the current pending service command request that the second acquisition module is got and processes.
9. the control system of service-oriented device cluster as claimed in claim 7, is characterized in that, described system also comprises:
The 3rd acquisition module, for obtaining the present load saturation of described server cluster, described load saturation degree is the ratio of the service command number of request of the current startup of described server cluster and the service command request disposal ability of described server cluster unit interval;
The first judge module, for judging whether described load saturation degree is greater than the first predetermined threshold value;
The first generation module, is connected with the first judge module, if described load saturation degree is greater than the first predetermined threshold value, according to the difference of described load saturation degree and the first predetermined threshold value, calculates the number of servers that described server cluster need to increase and generates lastest imformation.
10. the control system of service-oriented device cluster as claimed in claim 7, is characterized in that, described system also comprises:
The 4th acquisition module, for obtaining the present load saturation of described server cluster;
The second judge module, for judging whether described load saturation degree is less than the second predetermined threshold value;
The second generation module, is connected with the second judge module, if described load saturation degree is less than the second predetermined threshold value, according to the difference of described load saturation degree and the second predetermined threshold value, calculates and needs the number of servers of described server cluster minimizing and generate lastest imformation.
The control system of 11. service-oriented device clusters as claimed in claim 7, is characterized in that, described system also comprises:
Forwarding module, if for judging that the load saturation degree of first server is greater than the predetermined upper limit, issue other the effective server process in described server cluster by the one or more service command requests that start in described first server; Wherein, described first server is arbitrary effective server in described server cluster.
The control system of 12. service-oriented device clusters as claimed in claim 8, is characterized in that, in the processing module of described system, also comprises:
Priority block, for starting the minimum effective Servers installed priority of service command number of request to described service list;
Module is set, for effective Servers installed of judging described priority maximum as described processing server.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410439567.9A CN104158911A (en) | 2014-09-01 | 2014-09-01 | Server cluster-oriented control method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410439567.9A CN104158911A (en) | 2014-09-01 | 2014-09-01 | Server cluster-oriented control method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104158911A true CN104158911A (en) | 2014-11-19 |
Family
ID=51884316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410439567.9A Pending CN104158911A (en) | 2014-09-01 | 2014-09-01 | Server cluster-oriented control method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104158911A (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105162859A (en) * | 2015-08-20 | 2015-12-16 | 湖南亿谷科技发展股份有限公司 | Dynamic server dilatation system and method |
| CN105743772A (en) * | 2016-01-26 | 2016-07-06 | 深圳宸睿科技有限公司 | Message processing method and system |
| CN106357747A (en) * | 2016-08-30 | 2017-01-25 | 维沃移动通信有限公司 | Server load management method and device |
| CN106375395A (en) * | 2016-08-30 | 2017-02-01 | 厦门中学西渐信息科技有限公司 | Load balancing method and system for node servers |
| CN106789853A (en) * | 2015-11-24 | 2017-05-31 | 中国移动通信集团公司 | The dynamic dispatching method and device of a kind of transcoder |
| CN107040600A (en) * | 2017-04-20 | 2017-08-11 | 暴风集团股份有限公司 | A kind of real-time interactive method and real-time interactive system for competitive sports |
| CN107172156A (en) * | 2017-05-22 | 2017-09-15 | 中铁程科技有限责任公司 | A kind of server system and the vehicles |
| CN107844376A (en) * | 2017-11-21 | 2018-03-27 | 北京星河星云信息技术有限公司 | Resource allocation method, computing system, medium and the server of computing system |
| CN108471387A (en) * | 2018-03-27 | 2018-08-31 | 中国农业银行股份有限公司 | A kind of daily record flow decentralized control method and system |
| CN108737140A (en) * | 2017-04-20 | 2018-11-02 | 国基电子（上海）有限公司 | A kind of firmware update and controller |
| WO2019104974A1 (en) * | 2017-11-30 | 2019-06-06 | 平安科技（深圳）有限公司 | Dubbo platform-based automatic server starting and stopping method , server, and storage medium |
| CN109862376A (en) * | 2019-02-28 | 2019-06-07 | 广州华多网络科技有限公司 | Live content jettison system, method, apparatus, listserv and storage medium |
| CN110134575A (en) * | 2019-04-26 | 2019-08-16 | 厦门网宿有限公司 | A kind of the service ability calculation method and device of server cluster |
| CN110266525A (en) * | 2019-06-04 | 2019-09-20 | 湖北五五互联科技有限公司 | CDN server quantity configuration method, equipment and computer readable storage medium |
| CN110445828A (en) * | 2019-06-14 | 2019-11-12 | 平安科技（深圳）有限公司 | A kind of data distribution formula processing method and its relevant device based on Redis |
| CN110489221A (en) * | 2019-08-16 | 2019-11-22 | 南京云帐房网络科技有限公司 | A kind of method and system of automation task Number dynamics adjustment computing resource |
| CN110535939A (en) * | 2019-08-29 | 2019-12-03 | 深圳前海环融联易信息科技服务有限公司 | A kind of service discovery and method for pre-emptively, device, computer equipment and storage medium |
| CN111277373A (en) * | 2020-01-14 | 2020-06-12 | 深圳震有科技股份有限公司 | Server switching control method, terminal and storage medium |
| CN111901389A (en) * | 2020-07-03 | 2020-11-06 | 北京达佳互联信息技术有限公司 | Data updating method, device, server and storage medium |
| CN112769922A (en) * | 2020-12-31 | 2021-05-07 | 南京视察者智能科技有限公司 | Device and method for self-starting micro-service cluster |
| CN112783729A (en) * | 2021-01-29 | 2021-05-11 | 北京三快在线科技有限公司 | Exception handling method and exception handling device for gray scale release |
| CN113515357A (en) * | 2021-04-20 | 2021-10-19 | 建信金融科技有限责任公司 | Batch task execution method and device in cluster environment |
| CN113946376A (en) * | 2021-10-29 | 2022-01-18 | 挂号网（杭州）科技有限公司 | Load adjustment method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101957861A (en) * | 2010-10-18 | 2011-01-26 | 江苏大学 | Novel metadata server cluster and metadata management method based on reconciliation statement |
| CN102833355A (en) * | 2012-09-22 | 2012-12-19 | 广东电子工业研究院有限公司 | Load balance system and cloud computing oriented mechanism |
| CN103118076A (en) * | 2013-01-11 | 2013-05-22 | 烽火通信科技股份有限公司 | Upgraded server cluster system and load balancing method thereof |
| CN103490956A (en) * | 2013-09-22 | 2014-01-01 | 杭州华为数字技术有限公司 | Self-adaptive energy-saving control method, device and system based on traffic predication |
| CN103957246B (en) * | 2014-04-22 | 2017-03-01 | 广州杰赛科技股份有限公司 | Dynamic load balancing method based on tenant's perception and system |
-
2014
- 2014-09-01 CN CN201410439567.9A patent/CN104158911A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101957861A (en) * | 2010-10-18 | 2011-01-26 | 江苏大学 | Novel metadata server cluster and metadata management method based on reconciliation statement |
| CN102833355A (en) * | 2012-09-22 | 2012-12-19 | 广东电子工业研究院有限公司 | Load balance system and cloud computing oriented mechanism |
| CN103118076A (en) * | 2013-01-11 | 2013-05-22 | 烽火通信科技股份有限公司 | Upgraded server cluster system and load balancing method thereof |
| CN103490956A (en) * | 2013-09-22 | 2014-01-01 | 杭州华为数字技术有限公司 | Self-adaptive energy-saving control method, device and system based on traffic predication |
| CN103957246B (en) * | 2014-04-22 | 2017-03-01 | 广州杰赛科技股份有限公司 | Dynamic load balancing method based on tenant's perception and system |
Cited By (34)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105162859A (en) * | 2015-08-20 | 2015-12-16 | 湖南亿谷科技发展股份有限公司 | Dynamic server dilatation system and method |
| CN106789853A (en) * | 2015-11-24 | 2017-05-31 | 中国移动通信集团公司 | The dynamic dispatching method and device of a kind of transcoder |
| CN105743772A (en) * | 2016-01-26 | 2016-07-06 | 深圳宸睿科技有限公司 | Message processing method and system |
| CN106375395B (en) * | 2016-08-30 | 2019-06-11 | 厦门中学西渐信息科技有限公司 | The load-balancing method and system of node server |
| CN106357747A (en) * | 2016-08-30 | 2017-01-25 | 维沃移动通信有限公司 | Server load management method and device |
| CN106375395A (en) * | 2016-08-30 | 2017-02-01 | 厦门中学西渐信息科技有限公司 | Load balancing method and system for node servers |
| CN107040600A (en) * | 2017-04-20 | 2017-08-11 | 暴风集团股份有限公司 | A kind of real-time interactive method and real-time interactive system for competitive sports |
| CN108737140A (en) * | 2017-04-20 | 2018-11-02 | 国基电子（上海）有限公司 | A kind of firmware update and controller |
| CN108737140B (en) * | 2017-04-20 | 2021-05-28 | 国基电子（上海）有限公司 | Firmware updating method and controller |
| CN107172156A (en) * | 2017-05-22 | 2017-09-15 | 中铁程科技有限责任公司 | A kind of server system and the vehicles |
| CN107844376A (en) * | 2017-11-21 | 2018-03-27 | 北京星河星云信息技术有限公司 | Resource allocation method, computing system, medium and the server of computing system |
| WO2019104974A1 (en) * | 2017-11-30 | 2019-06-06 | 平安科技（深圳）有限公司 | Dubbo platform-based automatic server starting and stopping method , server, and storage medium |
| CN108471387A (en) * | 2018-03-27 | 2018-08-31 | 中国农业银行股份有限公司 | A kind of daily record flow decentralized control method and system |
| CN108471387B (en) * | 2018-03-27 | 2022-10-21 | 中国农业银行股份有限公司 | Log flow decentralized control method and system |
| CN109862376A (en) * | 2019-02-28 | 2019-06-07 | 广州华多网络科技有限公司 | Live content jettison system, method, apparatus, listserv and storage medium |
| CN109862376B (en) * | 2019-02-28 | 2021-06-01 | 广州方硅信息技术有限公司 | Live content delivery system, method and device, list server and storage medium |
| CN110134575A (en) * | 2019-04-26 | 2019-08-16 | 厦门网宿有限公司 | A kind of the service ability calculation method and device of server cluster |
| CN110134575B (en) * | 2019-04-26 | 2022-11-22 | 厦门网宿有限公司 | Method and device for calculating service capacity of server cluster |
| CN110266525A (en) * | 2019-06-04 | 2019-09-20 | 湖北五五互联科技有限公司 | CDN server quantity configuration method, equipment and computer readable storage medium |
| CN110266525B (en) * | 2019-06-04 | 2022-02-22 | 湖北五五互联科技有限公司 | CDN server number configuration method, equipment and computer readable storage medium |
| CN110445828A (en) * | 2019-06-14 | 2019-11-12 | 平安科技（深圳）有限公司 | A kind of data distribution formula processing method and its relevant device based on Redis |
| CN110445828B (en) * | 2019-06-14 | 2023-04-18 | 平安科技（深圳）有限公司 | Data distributed processing method based on Redis and related equipment thereof |
| CN110489221A (en) * | 2019-08-16 | 2019-11-22 | 南京云帐房网络科技有限公司 | A kind of method and system of automation task Number dynamics adjustment computing resource |
| CN110489221B (en) * | 2019-08-16 | 2023-10-20 | 云帐房网络科技有限公司 | Method and system for dynamically adjusting computing resources by automatic task number |
| CN110535939B (en) * | 2019-08-29 | 2022-02-11 | 深圳前海环融联易信息科技服务有限公司 | Service discovery and preemption method, device, computer equipment and storage medium |
| CN110535939A (en) * | 2019-08-29 | 2019-12-03 | 深圳前海环融联易信息科技服务有限公司 | A kind of service discovery and method for pre-emptively, device, computer equipment and storage medium |
| CN111277373A (en) * | 2020-01-14 | 2020-06-12 | 深圳震有科技股份有限公司 | Server switching control method, terminal and storage medium |
| CN111901389A (en) * | 2020-07-03 | 2020-11-06 | 北京达佳互联信息技术有限公司 | Data updating method, device, server and storage medium |
| CN111901389B (en) * | 2020-07-03 | 2023-07-04 | 北京达佳互联信息技术有限公司 | Data updating method, device, server and storage medium |
| CN112769922A (en) * | 2020-12-31 | 2021-05-07 | 南京视察者智能科技有限公司 | Device and method for self-starting micro-service cluster |
| CN112783729A (en) * | 2021-01-29 | 2021-05-11 | 北京三快在线科技有限公司 | Exception handling method and exception handling device for gray scale release |
| CN113515357A (en) * | 2021-04-20 | 2021-10-19 | 建信金融科技有限责任公司 | Batch task execution method and device in cluster environment |
| CN113515357B (en) * | 2021-04-20 | 2024-03-08 | 建信金融科技有限责任公司 | Method and device for executing batch tasks in cluster environment |
| CN113946376A (en) * | 2021-10-29 | 2022-01-18 | 挂号网（杭州）科技有限公司 | Load adjustment method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104158911A (en) | Server cluster-oriented control method and system | |
| CN102929707B (en) | Parallel task dynamical allocation method | |
| CN101354664B (en) | Method and apparatus for interrupting load equilibrium of multi-core processor | |
| CN109040212A (en) | Equipment access server cluster method, system, equipment and storage medium | |
| US9110696B2 (en) | Thin client system, connection management server, connection management method and connection management program | |
| CN102158513A (en) | Service cluster and energy-saving method and device thereof | |
| CN103516744A (en) | A data processing method, an application server and an application server cluster | |
| CN104679594B (en) | A kind of middleware distributed computing method | |
| EP3575979B1 (en) | Query priority and operation-aware communication buffer management | |
| CN103164253A (en) | Virtual machine deployment system and virtual machine deployment method | |
| CN104239227A (en) | Virtualization storage management device and local virtualization storage management device and method | |
| CN107864211A (en) | Cluster resource dispatching method and system | |
| CN105516325A (en) | Cloud load balancing method for carrying out elastic expansion and traffic distribution expansion according to application load | |
| EP3710306B1 (en) | Controlling electric vehicle charging currents | |
| CN105450684A (en) | Cloud computing resource scheduling method and system | |
| CN106412030B (en) | A kind of selection storage resource method, apparatus and system | |
| CN113342477A (en) | Container group deployment method, device, equipment and storage medium | |
| CN105406983A (en) | Cloud calculation network management control method based on user demands | |
| KR20130130295A (en) | System and method for assigining server to terminal and efficiently delivering messages to the terminal | |
| CN110309229A (en) | The data processing method and distributed system of distributed system | |
| CN112019577A (en) | Exclusive cloud storage implementation method and device, computing equipment and computer storage medium | |
| CN105245367A (en) | Serial number generation method and electronic device | |
| CN104778125A (en) | Method and system for memory management | |
| CN114744686B (en) | Generator set identification method, device, equipment and storage medium | |
| CN104812002B (en) | Process resource dynamic adjusting method, equipment and system under centralized RAN framework |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141119 |