CN105591964B - A kind of overload protection arrangement and method for internet system - Google Patents
A kind of overload protection arrangement and method for internet system Download PDFInfo
- Publication number
- CN105591964B CN105591964B CN201510665499.2A CN201510665499A CN105591964B CN 105591964 B CN105591964 B CN 105591964B CN 201510665499 A CN201510665499 A CN 201510665499A CN 105591964 B CN105591964 B CN 105591964B
- Authority
- CN
- China
- Prior art keywords
- request
- backend application
- client
- backend
- load
- 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.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
Abstract
Overload protection arrangement for internet system includes: acquisition module; the traffic mirroring of its network flow before reaching backend application acquires data; decision-making module; it determines the degree of load about the backend application according to the data; and determine Preservation tactics; protective module executes the Preservation tactics.
Description
Technical field
The embodiment of the present invention is related to network computer system load control, and more particularly relates to internet system
The overload protection arrangement and method of system.
Background technique
The introduction of this part, which can promote, more fully understands aspect of the invention.Correspondingly, the statement of this part should with regard to this and
By reading, and it is understood not to recognize that belongs to the prior art or what is not belonging to the prior art.
When facing the portfolio of sudden high pressure, the processing of network computer system especially Internet application system
Ability can be gradually reduced because of overload.The time that this will lead to the request of system processing business is elongated, and user is because losing patience not
Disconnected refresh page then further increases the pressure of system, causes avalanche effect.In this case, each of system processing is asked
It asks all by invalid due to time-out so that the service ability that system external portion is presented is almost 0.To avoid this experience, general place
Reason mode is to restart system.But do so the use for the service for providing system the processing for the business that directly affects and user.
Summary of the invention
Overload protection arrangement for internet system includes: acquisition module, the network before reaching backend application
The traffic mirroring of flow acquires data, and decision-making module determines the degree of load about the backend application according to the data, and determines
Preservation tactics, protective module execute the Preservation tactics.
Optionally, protective module is configured to execute according to the Preservation tactics: receiving the request from client, judgement should
Whether request is overtime;If the request has timed, out, response is returned to the client, without forwarding to the backend application, such as
The fruit request has not timed out, then according to the degree of load calculated by decision-making module, judges whether the backend application is capable of handling the request;
If the backend application is capable of handling the request, the request is forwarded to the backend application, if the backend application can not be located
The request is managed, then postpones to forward the request.
Optionally, which is configured to determine the type of service of the network flow, protective module root according to the data
It is configured to execute according to the Preservation tactics: receiving the request from client, according to the type of service of the network flow, judgement should
Whether request is core business request, if the request is that core business is requested, forwards the request to high-performance channel, should
High-performance channel is made of the node of more than half in the backend application, should if the request is core business request
Request is transmitted to general channel, which is made of the node in addition to the node in the high-performance channel.
Optionally, the degree of load about the backend application which determines according to the data, is configured to execute:
When judging that the degree of load is more than threshold value, automatic explosion command is sent to cloud dispatch server, so that the cloud dispatch server
The extension of cloud operation platform is notified to generate the new backend application of the request available to execute from client.
Optionally, protective module is configured to execute according to the Preservation tactics: the request from client is received, if should
Request has not timed out, then according to the degree of load calculated by decision-making module, judges whether the backend application is capable of handling the request;If
The backend application is capable of handling the request, then forwards the request to the backend application, if the backend application can not handle this
Request, then the backend application into standby resources pond forwards the request, wherein being deployed in advance in the standby resources pond multiple
The backend application having been started up.
Optionally, the data include error rate, response time, number of request per second.
A kind of overload protection method for internet system includes: the stream of the network flow before reaching backend application
It measures mirror image and acquires data, determine the degree of load about the backend application according to the data, and determine Preservation tactics, execute the protection
Strategy.
Optionally, it is executed according to the Preservation tactics: receiving the request from client, judge whether the request is overtime;Such as
The fruit request has timed, out, then returns to response to the client, without being forwarded to the backend application, if the request has not timed out,
Then according to the degree of load, judge whether the backend application is capable of handling the request;If the backend application is capable of handling this and asks
It asks, then forwards the request to the backend application, if the backend application can not handle the request, postpone to forward the request.
Optionally, the type of service that the network flow is determined according to the data is executed according to the Preservation tactics: reception comes from
The request of client judges whether the request is core business request, if the request according to the type of service of the network flow
Be core business request, then forward the request to high-performance channel, the high-performance channel by the half in the backend application with
On node composition, if the request be core business request, forward the request to general channel, the generality channel
It is made of the node in addition to the node in the high-performance channel.
Optionally, it according to the degree of load of the backend application, executes: when judging that the degree of load is more than threshold value, to cloud tune
It spends server and sends automatic explosion command, so that cloud dispatch server notice cloud operation platform extension generates available to execute come from
The new backend application of the request of client.
Optionally, it is executed according to the Preservation tactics: receiving the request from client, if the request has not timed out, root
According to the degree of load, judge whether the backend application is capable of handling the request;If the backend application is capable of handling the request,
The request is forwarded to the backend application, if the backend application can not handle the request, the rear end into standby resources pond
Using the request is forwarded, wherein multiple backend applications having been started up are deployed in the standby resources pond in advance.
One or more embodiment of the invention is monitored using the bypass of mirror image flow, will not to the data packet of original transmitting
Generate any influence;By application application response error rate, application response time, using number of request per second more accurately
Judgement applies itself whether be in overload;And there is provided a variety of safeguard measures.The present invention have with next or
Multiple advantages: in the case where systems face sudden high pressure, system service ability will not fall to 0, but can be to outreach
Show the maximum processing capability of system;The time has been reserved for subsequent dilatation, emergency, service will not be interrupted;In sudden high pressure
After request restores normal, adjust automatically state;System can continually and steadily be run, and improve the use in sudden high voltage power
Family experience.
The other feature and advantage that also will be understood that the embodiment of the present invention when being read in conjunction with the figure and being described below, wherein attached
Figure shows the principle of the embodiment of the present invention by means of example.
Detailed description of the invention
By means of example, from features as discussed above, the above and other aspect of each embodiment of the invention,
Feature and benefit will be apparent.
Fig. 1 shows the schematic diagram of overload protection arrangement according to an embodiment of the present invention in systems.
Fig. 2 shows the flow charts according to an embodiment of the present invention for realizing delay forwarding and overtime advanced processing.
Fig. 3 shows illustrative sudden high pressure curve graph.
Fig. 4 shows the illustrative sudden high pressure curve graph Jing Guo peak clipping.
Fig. 5 shows the flow chart according to an embodiment of the present invention for realizing protection key business.
Fig. 6 shows the signal of the overload protection arrangement that extends automatically of realization resource according to an embodiment of the present invention in systems
Figure.
Fig. 7 shows the schematic diagram of the overload protection according to an embodiment of the present invention using standby resources pond.
Specific embodiment
Hereinafter, the principle of the present invention and purport are described into reference implementation example.It should be understood that the embodiment provided
The present invention and further the practice present invention are more fully understood just for the sake of those skilled in the art, rather than limit the present invention
Range.For example, a part as one embodiment is illustrated or the feature of description can be used in combination with another embodiment and
Generate another embodiment.For clarity, all features of actual implementation mode are not described in the present specification.It is aobvious
So, it is to be appreciated that in the practical research and development of any this embodiment, for the specific objective for realizing developer, such as meet
The relevant constraint of system constraint relevant with business, embodiment will be different.
Disclosed theme is described with reference to the drawings.Various structures, system and equipment are schematically depicted in figure,
This be only for purposes of illustration rather than by these description with well known to a person skilled in the art details to obscure.However, including attached
Figure is the illustrative example in order to describe and explain published subject.
Fig. 1 shows the schematic diagram of overload protection arrangement according to an embodiment of the present invention in systems.As shown in Figure 1, system
It is made of client and server end.Request of the received server-side from client, handles the request, and by the result of processing
Return to client.Server end is usually provided with the load balancing network equipment for controlling load balancing, Web using for locating
It manages page logic and backend application is used for processing business logic.Wherein, the request from client mainly by backend application into
Row calculates.Here, backend application can be multiple calculate nodes.The calculate node for having processing capacity can be with physical form
Existing individual server is also possible to be present in the virtual computing unit in server.Therefore server end includes one
Or multiple backend applications carry out processing business request.
The embodiment of the present invention carries out overload protection to system by the way that overload protection arrangement is arranged in server end.Such as Fig. 1
Shown, overload protection arrangement can be arranged on the load balancing network equipment, Web application or backend application, from visitor
The request at family end reaches be responsible for carrying out the practical backend application for calculating work before the request is pre-processed in order to avoid system is because of mistake
It carries and computing capability is caused to reduce.As shown, overload protection arrangement, the load balancing network equipment, Web application layer collectively form
Overload protection layer.Overload protection arrangement includes acquisition module, decision-making module, protective module.Optionally, regulation engine and flow mirror
As module can also be arranged in overload protection arrangement.
According to an embodiment of the invention, acquisition module is configured to the flow of the network flow before reaching backend application
Mirror image acquires data, and decision-making module is configured to determine the degree of load about the backend application according to the data, and determines protection
Strategy and protective module are configured to execute the Preservation tactics.
Acquisition module, decision-making module, protective module, regulation engine and traffic mirroring module is described below to protect in process
Interaction in journey.Firstly, the network flow before reaching backend application is mirrored to shunting environment by traffic mirroring module, as adopting
Collect the data source of module.It carries out bypassing the influence that monitors and can be avoided to regular traffic processing by generating traffic mirroring.However,
Data are acquired from traffic mirroring by acquisition module, such as may include that the keys such as error rate, response time, number of request per second refer to
Mark.Then, decision-making module carries out the calculating of the degree of load of backend application according to the data that acquisition module provides, and can basis
The different loads configured in regulation engine strategy different when spending, notice protective module execute corresponding strategy.Finally, protection mould
For block after decision-making module reception notice, implementation strategy protects backend application.Protective module can be in load-balancing device
Or it is realized in a manner of F5+Apache or LVS+Nginx in WEB application.Protective module is executing decision-making module according to acquisition
Data generate strategy after, can execute again decision-making module according to think acquisition data generation strategy, be consequently formed
The positive dynamical feedback of one closure.
Delay forwarding and overtime advanced processing
Fig. 2 shows the flow charts according to an embodiment of the present invention for realizing delay forwarding and overtime advanced processing.Such as Fig. 2 institute
Show, the process the following steps are included:
In step 201, the request from client is received;
In step 202, judge whether the request is overtime;
If the request has timed, out, response is returned to the client in step 203, without turning to the backend application
Hair, thus the advanced processing request;
If the request has not timed out, in step 204 according to the degree of load calculated by decision-making module, the backend application is judged
Whether the processing capacity for whether being capable of handling the request or the backend application has reached the limit;
If the backend application is capable of handling the request, the request is forwarded to the backend application in step 206,
If the backend application can not handle the request, postpone to forward the request in step 205.
The strategy execution that above-mentioned steps can be generated by protective module according to decision-making module.Delay forwarding and time-out are located in advance
The request pressure that reason can ensure to be transmitted to backend application is no more than the maximum processing capability of backend application, and control is to the back-end
Using the request rate of forwarding to realize peak clipping effect, and the time for making backend application handle each request keeps steady
It is fixed.On the other hand, it for alreading exceed the request of client timeout time, is no longer forwarded toward backend application, i.e., in overload protection
Layer just carries out return response in advance, to reduce the pressure of backend application.
Fig. 3 and Fig. 4 is exemplarily illustrated the attainable mistake of delay forwarding according to Fig.2, and overtime advanced processing step institute
Carry the effect signal of protection.Fig. 3 shows illustrative sudden high pressure curve graph.Fig. 4 is shown illustratively by peak clipping
Sudden high pressure curve graph.Fig. 3 indicates the curve graph of the throughput of system TPS under sudden high pressure scene, horizontal dotted line
The maximum processing capability for indicating backend application, when TPS curve is more than horizontal dotted line, backend application will appear the abnormal feelings of overload
Condition.It is forwarded by delay, the request indicated with inclination solid line in Fig. 3 more than horizontal dotted line, can be moved to inclination solid line indicates
Part.So as to realize TPS curve graph as shown in Figure 4, wherein TPS curve is below horizontal dotted line.
Protect key business
Fig. 5 shows the flow chart according to an embodiment of the present invention for realizing protection key business.As shown in figure 5, in step
In 501, judge whether request is core business request;If the request is core business request, this is asked in step 502
It asks and is transmitted to high-performance channel;If the request is core business request, generality is forwarded the request in step 504
Channel.High-performance channel is made of most of node in the backend application, the remaining fraction node group in backend application
At general channel.For example, high-performance channel is made of the node of more than half in the backend application, the generality channel by
Node composition in addition to the node in the high-performance channel.It optionally, can also be by decision-making module according to from acquisition mould
The data of block dynamically divide high-performance channel and general channel.For example, decision-making module can be according to the current of each node
Processing capacity dynamically divides high-performance channel and general channel.Overload protection arrangement as shown in connection with fig. 1, the decision model
Block is configured to determine the type of service of the network flow according to the data from acquisition module, and protective module is according to the Preservation tactics
It is configured to execute: receiving the request from client, the business is forwarded according to the type of service of the network flow.As a result,
It may insure that core business request is preferentially quickly handled;The service request of non-critical path is queued processing.
Resource extends automatically
Fig. 6 shows the overload protection arrangement that extends automatically of realization resource according to an embodiment of the present invention showing in systems
It is intended to.As shown in fig. 6, the degree of load about the backend application that decision-making module can determine according to the collected data, is configured
At execution: when judging that the degree of load is more than threshold value, automatic explosion command is sent to cloud dispatch server, so that the cloud is dispatched
Server notifies the extension of cloud operation platform to generate the new backend application of the request available to execute from client.Cloud operation platform
According to the automatic extension backend application of notice from cloud dispatch server, the new application node of automatic pull-up.It is completed in extension
Afterwards, cloud dispatch server reinforms decision-making module.Decision-making module can configure overload protection according to the concrete condition of point spread
Strategy, to execute new Preservation tactics by protective module.For example, protective module is subsequent to be carried out to the node application newly extended
Request forwarding.
Standby resources pond
Fig. 7 shows the schematic diagram of the overload protection according to an embodiment of the present invention using standby resources pond.As shown in fig. 7,
Dispose multiple backend applications in advance using standby resources pond, and start backend application, so when system have overload risk when without
Real-time extension need to be carried out again.Standby resources pond can improve resource utilization by multiple and different Application shares.For example, when one
After the degree of load of backend application reaches threshold value, being scheduled from external request, it is corresponding in standby resources pond to be dispatched to
Backend application.It is a step example using resource standby pool below,
1. overload protection layer receives the request for applying A to the back-end;
2. judging whether backend application A has reached the limit of processing capacity;
3. if not up to the limit, rear end are forwarded using A;
4. if applying A being forwarded into standby resources pond have reached the limit;
5. returning to step 1 continues with subsequent request.
In conjunction with the overload protection arrangement in Fig. 1 or Fig. 6, protective module can be configured to execute according to Preservation tactics:
The request from client is received,
If the request has not timed out, according to the degree of load calculated by decision-making module, judge that the backend application whether can
Handle the request;
If the backend application is capable of handling the request, the request is forwarded to the backend application,
If the backend application can not handle the request, the backend application into standby resources pond forwards the request,
It is wherein deployed with multiple backend applications having been started up in advance in the standby resources pond.
Generally, various exemplary embodiments can be implemented in software.It can be by controller, microprocessor or other calculating
Device executes above-mentioned software, although the invention is not limited thereto.Although the various aspects of exemplary embodiment of the present invention can quilt
It is shown and described as block diagram, flow chart or shows and describe using certain other graphical representations, but be well understood that, here
These blocks of description, device, system can be in hardware, software, firmware, special circuit or logic electricity as non-limiting examples
Implement in road, common hardware or controller or other computing devices or its certain combination.
Fig. 2, each frame shown in fig. 5 can be considered as method and step, and/or be considered as due to running computer program code
Caused by operate, and/or be considered as being configured to implementing the logic circuit component of multiple couplings of correlation function (several functions).
Although this specification includes many specific implementation details, these details are not necessarily to be construed as to any invention
The limitation of range or the range that may be claimed, but should be considered as can be specific to the specific embodiment of specific invention
The description of feature.The certain features described in the present specification in the context of respective embodiment can also be combined in single reality
It applies in example and implements.On the contrary, the various features described in the context of single embodiment can also respectively in various embodiments
Or implement in any suitable sub-portfolio.In addition, although feature can be described above as with specific combination and even
It plays a role as initially being advocated, but can be in certain feelings from combined one or more features claimed
It is removed from the combination under condition, and combination claimed can be for sub-portfolio or the variation of sub-portfolio.
Similarly, although operation is depicted in figure in a specific sequence, this is understood not to require according to shown
Particular order or execute these operations in sequential order, or the operation of all illustrations is required to be performed, to reach ideal
As a result.In some cases, multi-task parallel processing may be advantageous.In addition, each germline in embodiments thereof described above
The separation of system component is understood not to require to carry out this separation in all of the embodiments illustrated, and should be understood that described
Program assembly and system generally can be integrated one and play in single software product or be packaged into multiple software product.
It is understood, therefore, that the embodiment of the present invention is not limited to disclosed specific embodiment, and variation and
Other embodiments are intended within the scope of the appended claims.
Claims (6)
1. a kind of overload protection arrangement for internet system characterized by comprising
The traffic mirroring of acquisition module, the network flow before reaching backend application acquires data,
Decision-making module determines the degree of load about the backend application, and determining Preservation tactics according to the data,
Protective module executes the Preservation tactics,
Wherein, the data include error rate, response time, number of request per second,
Wherein, protective module is configured to execute according to the Preservation tactics:
The request from client is received,
Judge whether the request is overtime;
If the request has timed, out, response is returned to the client, without forwarding to the backend application, thus advanced processing
The request,
If the request has not timed out, according to the degree of load calculated by decision-making module, judge whether the backend application is capable of handling
The request;
If the backend application is capable of handling the request, the request is forwarded to the backend application,
If the backend application can not handle the request, postpone to forward the request,
Wherein, the degree of load about the backend application which determines according to the data, is configured to execute:
When judging that the degree of load is more than threshold value, automatic explosion command is sent to cloud dispatch server, so that cloud scheduling clothes
Device notice cloud operation platform of being engaged in extends the new backend application for generating the request available to execute from client.
2. overload protection arrangement as described in claim 1, which is characterized in that
The decision-making module is configured to determine the type of service of the network flow according to the data,
Protective module is configured to execute according to the Preservation tactics:
The request from client is received,
According to the type of service of the network flow, judge whether the request is core business request,
If the request is core business request, high-performance channel is forwarded the request to, the high-performance channel is by the rear end
The node of more than half composition in,
If the request is not core business request, general channel is forwarded the request to, the generality channel is by removing institute
State the node composition other than the node in high-performance channel.
3. overload protection arrangement as described in claim 1, which is characterized in that
Protective module is configured to execute according to the Preservation tactics:
The request from client is received,
If the request has not timed out, according to the degree of load calculated by decision-making module, judge whether the backend application is capable of handling
The request;
If the backend application is capable of handling the request, the request is forwarded to the backend application,
If the backend application can not handle the request, the backend application into standby resources pond forwards the request, wherein
Multiple backend applications having been started up are deployed in the standby resources pond in advance.
4. a kind of overload protection method for internet system characterized by comprising
The traffic mirroring of network flow before reaching backend application acquires data,
The degree of load about the backend application is determined according to the data, and determines Preservation tactics,
The Preservation tactics are executed,
Wherein, the data include error rate, response time, number of request per second, wherein
It is executed according to the Preservation tactics:
The request from client is received,
Judge whether the request is overtime;
If the request has timed, out, response is returned to the client, without forwarding to the backend application, thus advanced processing
The request,
If the request has not timed out, according to the degree of load, judge whether the backend application is capable of handling the request;
If the backend application is capable of handling the request, the request is forwarded to the backend application,
If the backend application can not handle the request, postpone to forward the request,
Wherein, it according to the degree of load of the backend application, executes:
When judging that the degree of load is more than threshold value, automatic explosion command is sent to cloud dispatch server, so that cloud scheduling clothes
Device notice cloud operation platform of being engaged in extends the new backend application for generating the request available to execute from client.
5. overload protection method as claimed in claim 4, which is characterized in that
The type of service of the network flow is determined according to the data,
It is executed according to the Preservation tactics:
The request from client is received,
According to the type of service of the network flow, judge whether the request is core business request,
If the request is core business request, high-performance channel is forwarded the request to, the high-performance channel is by the rear end
The node of more than half composition in,
If the request is not core business request, general channel is forwarded the request to, the generality channel is by removing institute
State the node composition other than the node in high-performance channel.
6. overload protection method as claimed in claim 4, which is characterized in that
It is executed according to the Preservation tactics:
The request from client is received,
If the request has not timed out, according to the degree of load, judge whether the backend application is capable of handling the request;
If the backend application is capable of handling the request, the request is forwarded to the backend application,
If the backend application can not handle the request, the backend application into standby resources pond forwards the request, wherein
Multiple backend applications having been started up are deployed in the standby resources pond in advance.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510665499.2A CN105591964B (en) | 2015-10-14 | 2015-10-14 | A kind of overload protection arrangement and method for internet system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510665499.2A CN105591964B (en) | 2015-10-14 | 2015-10-14 | A kind of overload protection arrangement and method for internet system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105591964A CN105591964A (en) | 2016-05-18 |
CN105591964B true CN105591964B (en) | 2019-08-13 |
Family
ID=55931196
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510665499.2A Active CN105591964B (en) | 2015-10-14 | 2015-10-14 | A kind of overload protection arrangement and method for internet system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105591964B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108206841B (en) * | 2016-12-16 | 2021-05-14 | 北京酷我科技有限公司 | Data processing method and system |
CN109785606A (en) * | 2019-02-02 | 2019-05-21 | 中能瑞通(北京)科技有限公司 | A kind of power information acquisition system acquisition quality analysis method and device |
CN111447113B (en) * | 2020-03-25 | 2021-08-27 | 中国建设银行股份有限公司 | System monitoring method and device |
CN111814024B (en) * | 2020-08-14 | 2021-09-03 | 北京斗米优聘科技发展有限公司 | Distributed data acquisition method, system and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102123179A (en) * | 2011-03-28 | 2011-07-13 | 中国人民解放军国防科学技术大学 | Load balancing method and system applied to distributed application system |
CN102420867A (en) * | 2011-12-01 | 2012-04-18 | 浪潮电子信息产业股份有限公司 | Cluster storage entry resolution method based on real-time load balancing mechanism |
CN102811157A (en) * | 2011-06-01 | 2012-12-05 | 阿尔卡特朗讯公司 | Method and device for flow control |
CN103179217A (en) * | 2013-04-19 | 2013-06-26 | 中国建设银行股份有限公司 | Load balancing method and device applicable to WEB application server group |
-
2015
- 2015-10-14 CN CN201510665499.2A patent/CN105591964B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102123179A (en) * | 2011-03-28 | 2011-07-13 | 中国人民解放军国防科学技术大学 | Load balancing method and system applied to distributed application system |
CN102811157A (en) * | 2011-06-01 | 2012-12-05 | 阿尔卡特朗讯公司 | Method and device for flow control |
CN102420867A (en) * | 2011-12-01 | 2012-04-18 | 浪潮电子信息产业股份有限公司 | Cluster storage entry resolution method based on real-time load balancing mechanism |
CN103179217A (en) * | 2013-04-19 | 2013-06-26 | 中国建设银行股份有限公司 | Load balancing method and device applicable to WEB application server group |
Also Published As
Publication number | Publication date |
---|---|
CN105591964A (en) | 2016-05-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105591964B (en) | A kind of overload protection arrangement and method for internet system | |
JP6457447B2 (en) | Data center network traffic scheduling method and apparatus | |
CN105100184B (en) | Reliable and deterministic live migration of virtual machines | |
US10805385B2 (en) | Method and apparatus for managing resource on cloud platform | |
US9568975B2 (en) | Power balancing to increase workload density and improve energy efficiency | |
CN103365704B (en) | Internal memory pre-copy method in virtual machine (vm) migration and perform device and the system of the method | |
CN107003706A (en) | For when the power distribution to Distributed Computer System is changed management can be with the operation that can not be suspended method and apparatus | |
US8588094B2 (en) | Computer-readable medium storing test environment creation program, test environment creation method, and test device | |
CN107645520B (en) | Load balancing method, device and system | |
Panwar et al. | A comparative study of load balancing algorithms in cloud computing | |
WO2013158139A1 (en) | Virtual computing resource orchestration | |
CN107967164B (en) | Method and system for live migration of virtual machine | |
Tighe et al. | Integrating cloud application autoscaling with dynamic VM allocation | |
KR20170139872A (en) | Multi-tenant based system and method for providing services | |
RU2669004C2 (en) | Method and apparatus for controlling direct-current voltage | |
WO2017112324A1 (en) | Ranking system | |
Mandal et al. | Heterogeneous bandwidth provisioning for virtual machine migration over SDN-enabled optical networks | |
CN106059940B (en) | A kind of flow control methods and device | |
CN105487928A (en) | Control method and device and Hadoop system | |
KR101608859B1 (en) | Cloud based intelligent power control system | |
CN109739634A (en) | A kind of atomic task execution method and device | |
Sakthivelmurugan et al. | Enhanced load balancing technique in public cloud | |
JP6186287B2 (en) | System management server and control method | |
JP5876425B2 (en) | Virtual machine live migration scheduling method and virtual machine system | |
KR101758065B1 (en) | Load balancing system, method for controlling the same and recording medium for performing the ethod |
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 |