CN103379168A - Data center resource distribution management method and system - Google Patents
Data center resource distribution management method and system Download PDFInfo
- Publication number
- CN103379168A CN103379168A CN2012101302705A CN201210130270A CN103379168A CN 103379168 A CN103379168 A CN 103379168A CN 2012101302705 A CN2012101302705 A CN 2012101302705A CN 201210130270 A CN201210130270 A CN 201210130270A CN 103379168 A CN103379168 A CN 103379168A
- Authority
- CN
- China
- Prior art keywords
- user
- server
- resource
- virtual machine
- execution
- 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
Images
Abstract
The invention discloses a data center resource distribution management method and system, and relates to the technical field of management and distribution of a data center system and system resources. The data center resource distribution management method comprises the following steps that user requirements are collected, wherein each user requirement comprises a list, applied by the user, of a virtual machine; resource information of a data center hardware system is collected, and the resource information of the system and the current state of each server are fed back in real time; the index of the server which the virtual machine applied by the user should be deployed is calculated according to the user requirements and the resource information of a current system, and the construction of the virtual machine applied by the user in the corresponding driver is conducted according to the index. The data center resource distribution management method and system can ensure fairness and reliability of a distributing result and enable the utilization rate of system resources to be improved in a certain extent.
Description
Technical field
The present invention relates to management and the distribution technique field of data center systems and system resource, relate in particular to a kind of data center resource allocation management method and system.
Background technology
Data center be on the Internet various services the center is provided, gathered the different classes of resources such as a large amount of computings, storage in the data center, need to reasonably be assigned in each user's hand.Different user's emphasis is different, and is also not identical to the demand of resource, and resource is distributed should be able to guarantee that allocation result is fair and reasonable, and each user's primary demand is guaranteed.
The energy resource consumption of data center is a huge problem, be Internet resources take or the expenditure of electric power resource all like this.Resource is distributed the effect should be able to reach energy savings, and this need to share out the work as far as possible so that resource utilization effectively improves, avoid a large amount of equipment or in long-time idling carry or hang down load running.
Summary of the invention
The technical problem that (one) will solve
The technical problem to be solved in the present invention is: provides a kind of and can guarantee the fair reliable of allocation result, and also can be to a certain extent so that data center's resource allocation management method and system that the utilance of system resource promotes.
(2) technical scheme
For addressing the above problem, the invention provides a kind of data center resource allocation management method, the method comprising the steps of:
S1. gather user request, comprise the tabulation of the virtual machine that described user applies in described user's request;
The resource information of the S2. resource information of image data center hardware system, and real-time feedback system and the state of current each server;
S3. according to the resource information of described user's request and current system, the index of the server that the virtual machine that the calculating user applies for should be deployed in is set up the virtual machine that the user applies for according to described index execution at corresponding server.
Preferably, the resource information of described hardware system comprises CPU usage that each server is current and availability, the free space of storage system, the internal network transmittability is estimated and external network transmission available bandwidth information.
Preferably, the state of described server comprises the current virtual machine quantity that has disposed.
Preferably, in step S3, also comprise step: provide the interface that connects its virtual machine of applying for to the user.
Preferably, among the step S3, the computational methods of the index of the server that the virtual machine that the user applies for should be deployed in comprise step:
The current resource information that can provide of server is provided A1;
If the described resource of A2 satisfies the resource of user's needs, execution in step A5 then, otherwise execution in step A3;
If A3 has traveled through Servers-all, execution in step A6 then, otherwise execution in step A4;
A4 reads the resource that next station server can provide, and returns steps A 2;
A5 exports the sign of the server of the resource that satisfies user's needs;
A6 returns the prompting that can't satisfy user's needs to the user, and carries out mistake and process.
Preferably, among the step S3, the computational methods of the index of the server that the virtual machine that the user applies for should be deployed in comprise step:
B1 reads server S
xThe current resource that can provide
And calculate the new array of its n-1 unit
If the described resource of B2
Satisfy the resource (R of user's needs
1, R
2, R
3... R
n), execution in step B5 then, otherwise execution in step B3;
If B3 has traveled through Servers-all, execution in step B6 then, otherwise execution in step B4;
B4 reads next station server S
X+1The resource that can provide
Calculate its new array of n-1 unit, and return step B2;
The B5 calculating target function, the calculated value and the setting threshold that compare the current goal function, if the calculated value of described current goal function is less than described setting threshold, then give described setting threshold with the calculated value assignment of described current goal function, the sign of record current server, and return step B3, if the calculated value of described current goal function equals described setting threshold, execution in step B7 then, if the calculated value of described current goal function greater than described setting threshold, execution in step B6 then;
B6 exports the sign of the server of the resource that satisfies user's needs;
B7 returns the prompting that can't meet consumers' demand to the user, and carries out mistake and process;
Wherein, (R
1, R
2, R
3... R
n) be the n-tuple group, the stock number that expression is applied for n different resource; M is the number of units of server, x=1, and 2 ..., m.
Preferably, described setting threshold is 10000000.
Preferably, described target function is:
The present invention also provides resource allocation management system of a kind of data center, comprising: acquisition module, be used for to gather user's request, and comprise the tabulation of the virtual machine that described user applies in described user's request; Feedback module, for the resource information of image data center hardware system, and the state of the resource information of real-time feedback system and current each server; Executive Module, according to the resource information of described user's request and current system, the index of the server that the virtual machine that the calculating user applies for should be deployed in is set up the virtual machine that the user applies for according to described index execution at corresponding server.
Preferably, described Executive Module also is used for providing the interface that connects its virtual machine of applying for to the user.
(3) beneficial effect
Data center of the present invention resource allocation management method and system provide a cover complete resource allocation management mechanism, can average out between the two in fair allocat and efficient allocation, both guaranteed the fair reliable of allocation result, also to a certain extent so that the utilance of resource be improved, thereby reach the effect of saving data center's energy.
Description of drawings
Fig. 1 is the data center's resource allocation management Method And Principle schematic diagram according to one embodiment of the present invention;
Fig. 2 is the flow chart according to data center's resource allocation management method of one embodiment of the present invention;
The flow chart of the method for the index of the server that Fig. 3 should be deployed in for the virtual machine that calculates the user according to the first in data center's resource allocation management method of one embodiment of the present invention and ask;
The flow chart of the method for the index of the server that Fig. 4 should be deployed in for the virtual machine that calculates the user according to the second in data center's resource allocation management method of one embodiment of the present invention and ask.
Embodiment
Data center's resource allocation management method and system that the present invention proposes reach by reference to the accompanying drawings embodiment and are described in detail as follows.
The resource allocation management framework of data center as shown in Figure 1, data center's resource allocation management method according to one embodiment of the present invention is the pith that moves in the cloud management platform (data center's management platform), can dispose on the management platform that is integrated into data center, under the prerequisite of the available resources information that obtains data center and user's solicited message, the resource at data center is managed and distributes.The mode that realizes is to utilize virtual machine technique, the resource in the resource pool divided and made up, thus the distribution of realization customizable.As shown in Figure 2, the method comprising the steps of:
S1. gather user's request.Generally speaking, the user applies in the usage data that resource in the heart is to realize by the application virtual machine.A user asks generally to comprise a series of virtual machine tabulation, and every virtual machine is configured demarcation by the user on the basis of understanding system resource and resource price, such as core cpu quantity, and running memory size and Internet Transmission Bandwidth guaranteed etc.
The resource information of the S2. resource information of image data center hardware system, and real-time feedback system and the state of current each server.The resource information of hardware system comprises CPU usage that each server is current and availability, the free space of storage system, the internal network transmittability is estimated and external network transmission available bandwidth etc. information, these information will be recorded and constantly update, thereby guarantee the carrying out of distribution.The state of current server comprises current virtual machine quantity that has disposed of this server etc., and these information feed back on the one hand the manager and are used for Real Time Monitoring, also feed back on the other hand the user, as the reference of resource request.
S3. according to the resource information of user request and current system, the index of the server that the virtual machine that the calculating user applies for should be deployed in, set up the virtual machine that the user applies for according to this index execution at corresponding server, and provide the interface that connects its virtual machine of applying for to the user.
Suppose that the resource that the user needs is array (R
1, R
2, R
3... R
n), the n-tuple group represents n different resource requested resource amount; M station server S
1, S
2, S
3... S
mIn any station server S
xThe resource that can provide is the n-tuple group
X=1,2 ..., m.
In the rapid S3 of the method for present embodiment, the computational methods of the index of the server that the virtual machine that the user applies for should be deployed in comprise following two kinds of methods:
First method is paid attention to dispensing rate, as shown in Figure 3, comprises step:
If A2 is S
xThe current resource that can provide
Satisfy the resource (R of user's needs
1, R
2, R
3... R
n), execution in step A5 then, otherwise execution in step A3;
If A3 has traveled through Servers-all, execution in step A6 then, otherwise execution in step A4;
A5 exports the sign of the server of the resource that satisfies user's needs;
A6 returns the prompting that can't satisfy user's needs to the user, and carries out mistake and process.
Second method is paid attention to distribution effects, as shown in Figure 4, comprises step:
B1 reads server S
xThe current resource that can provide
And calculate the new array of its n-1 unit
If B2 resource
Satisfy the resource (R of user's needs
1, R
2, R
3... R
n), execution in step B5 then, otherwise execution in step B3;
If B3 has traveled through Servers-all, execution in step B6 then, otherwise execution in step B4;
B4 reads next station server S
X+1The resource that can provide
Calculate its new array of n-1 unit, and return step B2;
The B5 calculating target function
The calculated value y and the setting threshold y that compare the current goal function
MinIf, y
Min>y, then with the calculated value y assignment of current goal function to setting threshold y
Min, the sign of record current server, and return step B3, if y
Min=y, if execution in step B7 then is y
Min<y, then execution in step B6.Target function can be according to different evaluation indexes, and different functions can be assessed the current allocation result that satisfies basic demand on the aspects such as different dimensions, angle, and evaluation index is then with actual resource distribution, user preference etc. factor analysis.What adopt in the present embodiment is the simplest target function setting.
B6 exports the sign of the server of the resource that satisfies user's needs;
B7 returns the prompting that can't meet consumers' demand to the user, carries out mistake and processes.
Setting threshold y
MinThe initial value setting to decide according to the distribution of target function, main purpose is in order to guarantee that the server that satisfies basic demand goes on record.Setting threshold y
MinShould be large number as far as possible, this be because: for example, if only have a station server, but its target function is very large, if this setting threshold is too little, such result can be dropped, thus cause the request unaccepted unfavorable phenomenon.In the method for present embodiment, y
Min=10000000.
The present invention also provides resource allocation management system of a kind of data center, and this system integration comprises in cloud management platform shown in Figure 1: acquisition module, be used for to gather user's request, and comprise the tabulation of the virtual machine that the user applies in user's request; Feedback module, for the resource information of image data center hardware system, and the state of the resource information of real-time feedback system and current each server; Executive Module, resource information according to user's request and current system, the index of the server that the virtual machine that the calculating user asks should be deployed in, set up the virtual machine that the user applies for according to the index execution at corresponding server, and provide the interface that connects its virtual machine of applying for to the user.
Above execution mode only is used for explanation the present invention; and be not limitation of the present invention; the those of ordinary skill in relevant technologies field; in the situation that does not break away from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.
Claims (10)
1. data center's resource allocation management method is characterized in that the method comprising the steps of:
S1. gather user request, comprise the tabulation of the virtual machine that described user applies in described user's request;
The resource information of the S2. resource information of image data center hardware system, and real-time feedback system and the state of current each server;
S3. according to the resource information of described user's request and current system, the index of the server that the virtual machine that the calculating user applies for should be deployed in is set up the virtual machine that the user applies for according to described index execution at corresponding server.
2. the method for claim 1, it is characterized in that the resource information of described hardware system comprises CPU usage that each server is current and availability, the free space of storage system, the internal network transmittability is estimated and external network transmission available bandwidth information.
3. the method for claim 1 is characterized in that, the state of described server comprises the current virtual machine quantity that has disposed.
4. the method for claim 1 is characterized in that, also comprises step in step S3: the interface that connects its virtual machine of applying for is provided to the user.
5. the method for claim 1 is characterized in that, among the step S3, the computational methods of the index of the server that the virtual machine that the user applies for should be deployed in comprise step:
The current resource information that can provide of server is provided A1;
If the described resource of A2 satisfies the resource of user's needs, execution in step A5 then, otherwise execution in step A3;
If A3 has traveled through Servers-all, execution in step A6 then, otherwise execution in step A4;
A4 reads the resource that next station server can provide, and returns steps A 2;
A5 exports the sign of the server of the resource that satisfies user's needs;
A6 returns the prompting that can't satisfy user's needs to the user, and carries out mistake and process.
6. the method for claim 1 is characterized in that, among the step S3, the computational methods of the index of the server that the virtual machine that the user applies for should be deployed in comprise step:
B1 reads server S
xThe current resource that can provide
And calculate the new array of its n-1 unit
If the described resource of B2
Satisfy the resource (R of user's needs
1, R
2, R
3... R
n), execution in step B5 then, otherwise execution in step B3;
If B3 has traveled through Servers-all, execution in step B6 then, otherwise execution in step B4;
B4 reads next station server S
X+1The resource that can provide
Calculate its new array of n-1 unit, and return step B2;
The B5 calculating target function, the calculated value and the setting threshold that compare the current goal function, if the calculated value of described current goal function is less than described setting threshold, then give described setting threshold with the calculated value assignment of described current goal function, the sign of record current server, and return step B3, if the calculated value of described current goal function equals described setting threshold, execution in step B7 then, if the calculated value of described current goal function greater than described setting threshold, execution in step B6 then;
B6 exports the sign of the server of the resource that satisfies user's needs;
B7 returns the prompting that can't meet consumers' demand to the user, and carries out mistake and process;
Wherein, (R
1, R
2, R
3... R
n) be the n-tuple group, the stock number that expression is applied for n different resource; M is the number of units of server, x=1, and 2 ..., m.
7. method as claimed in claim 6 is characterized in that, described setting threshold is 10000000.
8. method as claimed in claim 6 is characterized in that, described target function is:
9. a resource allocation management system of data center is characterized in that, comprising:
Acquisition module be used for to gather user's request, comprises the tabulation of the virtual machine that described user applies in described user's request;
Feedback module, for the resource information of image data center hardware system, and the state of the resource information of real-time feedback system and current each server;
Executive Module, according to the resource information of described user's request and current system, the index of the server that the virtual machine that the calculating user applies for should be deployed in is set up the virtual machine that the user applies for according to described index execution at corresponding server.
10. system as claimed in claim 9 is characterized in that, described Executive Module also is used for providing the interface that connects its virtual machine of applying for to the user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2012101302705A CN103379168A (en) | 2012-04-27 | 2012-04-27 | Data center resource distribution management method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2012101302705A CN103379168A (en) | 2012-04-27 | 2012-04-27 | Data center resource distribution management method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103379168A true CN103379168A (en) | 2013-10-30 |
Family
ID=49463724
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2012101302705A Pending CN103379168A (en) | 2012-04-27 | 2012-04-27 | Data center resource distribution management method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103379168A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103729254A (en) * | 2013-12-31 | 2014-04-16 | 国网山东省电力公司 | Virtual server Virtual CPU resource monitoring and dynamic allocation method |
CN104601585A (en) * | 2015-01-29 | 2015-05-06 | 武汉理工大学 | Multimedia cloud data center energy efficient resource allocation method based on user experiences |
CN105183561A (en) * | 2015-09-02 | 2015-12-23 | 浪潮(北京)电子信息产业有限公司 | Resource distribution method and resource distribution system |
CN108132839A (en) * | 2016-11-30 | 2018-06-08 | 中国移动通信有限公司研究院 | A kind of resource regulating method and device |
CN109302308A (en) * | 2018-08-23 | 2019-02-01 | 厦门秦淮科技有限公司 | A kind of data center IDC resource management system and management method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102270157A (en) * | 2011-07-25 | 2011-12-07 | 浪潮(北京)电子信息产业有限公司 | Server resource allocation system and method |
CN102307133A (en) * | 2011-03-25 | 2012-01-04 | 国云科技股份有限公司 | Virtual machine scheduling method for public cloud platform |
-
2012
- 2012-04-27 CN CN2012101302705A patent/CN103379168A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102307133A (en) * | 2011-03-25 | 2012-01-04 | 国云科技股份有限公司 | Virtual machine scheduling method for public cloud platform |
CN102270157A (en) * | 2011-07-25 | 2011-12-07 | 浪潮(北京)电子信息产业有限公司 | Server resource allocation system and method |
Non-Patent Citations (1)
Title |
---|
袁晓洁,: ""虚拟用户环境中资源动态管理机制研究"", 《中国知网-信息科技辑》, no. 2, 15 December 2011 (2011-12-15) * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103729254A (en) * | 2013-12-31 | 2014-04-16 | 国网山东省电力公司 | Virtual server Virtual CPU resource monitoring and dynamic allocation method |
CN104601585A (en) * | 2015-01-29 | 2015-05-06 | 武汉理工大学 | Multimedia cloud data center energy efficient resource allocation method based on user experiences |
CN104601585B (en) * | 2015-01-29 | 2018-12-14 | 武汉理工大学 | Multimedia cloud data center high energy efficiency resource allocation methods based on user experience |
CN105183561A (en) * | 2015-09-02 | 2015-12-23 | 浪潮(北京)电子信息产业有限公司 | Resource distribution method and resource distribution system |
CN105183561B (en) * | 2015-09-02 | 2018-09-14 | 浪潮(北京)电子信息产业有限公司 | A kind of resource allocation methods and system |
CN108132839A (en) * | 2016-11-30 | 2018-06-08 | 中国移动通信有限公司研究院 | A kind of resource regulating method and device |
CN108132839B (en) * | 2016-11-30 | 2020-10-30 | 中国移动通信有限公司研究院 | Resource scheduling method and device |
CN109302308A (en) * | 2018-08-23 | 2019-02-01 | 厦门秦淮科技有限公司 | A kind of data center IDC resource management system and management method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103020752B (en) | Automatic order distribution system | |
Li et al. | Optimal resource provisioning for cloud computing environment | |
CN104881325A (en) | Resource scheduling method and resource scheduling system | |
CN103763343A (en) | Method and device for processing service access | |
US20200012523A1 (en) | System and method for scheduling workload based on a credit-based mechanism | |
CN106506670B (en) | A kind of cloud platform virtual resource high speed dispatching method and system | |
CN103023963B (en) | A kind of method for cloud storage resources configuration optimization | |
CN103605567A (en) | Cloud computing task scheduling method facing real-time demand change | |
US20150058844A1 (en) | Virtual computing resource orchestration | |
CN104021040A (en) | Cloud computing associated task scheduling method and device based on time constraint | |
CN108337109A (en) | A kind of resource allocation methods and device and resource allocation system | |
CN103379168A (en) | Data center resource distribution management method and system | |
CN103180830A (en) | Cpu resource distribution method and calculating node | |
CN104142860A (en) | Resource adjusting method and device of application service system | |
US20110004500A1 (en) | Allocating a resource based on quality-of-service considerations | |
CN103581339A (en) | Storage resource allocation monitoring and processing method based on cloud computing | |
US20200059097A1 (en) | Providing Energy Elasticity Services Via Distributed Virtual Batteries | |
CN109684078A (en) | Resource dynamic distributing method and system for spark streaming | |
Xiao et al. | A priority based scheduling strategy for virtual machine allocations in cloud computing environment | |
CN104574051A (en) | Charging method and device | |
CN112395075A (en) | Resource processing method and device and resource scheduling system | |
US10379586B2 (en) | System and method for a data center to provide energy elasticity services at different time scales | |
CN103995749B (en) | The calculation task allocating method and system of cell cloud system | |
CN103713852A (en) | Method for processing information, service platform and electronic devices | |
CN106856441A (en) | VIM systems of selection and device in NFVO |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20131030 |
|
RJ01 | Rejection of invention patent application after publication |