CN104063282A - Management method, device and server for IaaS cloud variable scale resource pool - Google Patents
Management method, device and server for IaaS cloud variable scale resource pool Download PDFInfo
- Publication number
- CN104063282A CN104063282A CN201410188323.8A CN201410188323A CN104063282A CN 104063282 A CN104063282 A CN 104063282A CN 201410188323 A CN201410188323 A CN 201410188323A CN 104063282 A CN104063282 A CN 104063282A
- Authority
- CN
- China
- Prior art keywords
- resource pool
- physical machine
- task
- pool management
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a management method, device and server for an IaaS cloud variable scale resource pool. The method of the invention aims to trace the variation tendencies of the actual operating load of an IaaS system and physical computer resources; then, whether or not carrying out the operations of expanding or deleting for the resource pool are determined dynamically according to the prediction for frequentness of a future task; the operating cost of the system during loading is lowered under the condition that system performance during high loading is ensured as far as possible, so that optimization and balance are achieved, and both of the performance and the economical property of the cloud system can be considered.
Description
Technical field
The invention belongs to the technical field of cloud computing, the field that when especially facing cloud calculates operation, real-time scheduling of resource and control are optimized.
Background technology
Cloud computing refers to task distribution on a large amount of distributed computers, uses cloud computing platform, by network, provides the computation schema of information service for user.With respect to traditional software forms, the significant advantage such as cloud computing has loose couplings, on-demand, cost is controlled, resource is virtual, isomery is collaborative, makes it more adapt to the application such as ecommerce now, flexible manufacturing, mobile Internet.The implication that cloud computing comprises two aspects a: aspect is the cloud computing platform infrastructure that bottom builds, and is for building the basis of upper level applications; Implication is on the other hand the cloud computing application program being structured on this basic platform.Cloud computing can be divided three classes according to COS: using infrastructure as service (IaaS, Infrastructure asa Service), using platform as service (PaaS, Platform as a Service) and using software as service (SaaS, Software asa Service).
IaaS pattern cloud computing platform is that the Intel Virtualization Technologies such as, internal memory virtualization virtual by system virtualization, multiprocessor, I/O be virtual are virtualized into resource pool by physical resource, and these resources are carried out unified management and dispatching by cloud computing platform again.At present, have a lot of enterprises and scientific research institution to release the IaaS cloud computing platform of oneself, user oriented provides computational resource and storage resources.That the most representative is the elasticity calculating cloud EC2 (Elastic Compute Cloud) of Amazon (Amazon).
Main computational resource in IaaS cloud resource pool, be exactly physical machine (Physical Machine, PM), physical machine is the cloud task executing units the most basic (a physical machine cannot split into a plurality of muon physics machines again) that can not segment again, is the set of a certain amount of calculating, storage and network service resource.In a physical machine, can move one or more virtual machine processes, and a virtual machine process synchronization only may move in a physical machine.
In IaaS cloud system operational process, in the dynamic change that physical machine in resource pool always receives, carries out, discharges in task: if all physical machine in executing state have all reached the upper limit of the virtual machine number of processes that can support, by newly arrived virtual machine course allocation to idle physical machine; If all virtual machine processes of moving in a physical machine are all finished, this physical machine returns to idle condition.
Traditional cloud resource pool management technology, always safeguard a fixedly physical clusters for scale, there is shortcoming in this technology: because the physical machine quantity in resource pool is constant, when cloud system task load increases sharply, remaining idle physical machine will be had too many difficulties to cope with, and system performance is produced to adverse influence; Conversely, if cloud system, for a long time in extremely low load running, has a large amount of physical machine all in idle condition in resource pool, do not produce actual effectiveness; In resource pool, all physical machine is all when oepration at full load, and newly arrived task will be rejected, or is moved away, has affected greatly the execution efficiency of this task.
Under this background, the cloud resource pool management technology of scalable just becomes the solution of the problems referred to above.So-called scalable, in other words the physical machine quantity in cloud resource pool, can adjust dynamically according to the load variations of cloud system integral body: when expecting the trend of load increase, remaining idle physical machine may be difficult to deal with while needing future, expanding resource pond scale, increases idle physical machine quantity in advance; When expecting load downward trend, idle physical machine may be more time, be reduced resource pool scale following, reduces physical machine quantity (the idle physical machine of closed portion or the cloud system that the idle physical machine of part is transferred other to is used).Above-mentioned strategy, can guarantee the performance under the high load condition of system, can take into account again economical and energy saving and operation cost under low loading condition.
Yet, how to determine the control opportunity of above-mentioned operating strategy and increase and decrease quantity, guarantee the optimality of control strategy, be but a difficult problem.
Cloud computing itself belongs to an emerging technology field, and relevant technology, theory and method be also for the growth stage, the resource pool management technology of existing IaaS cloud, and there are the following problems:
(1) safeguard the resource pool of a fixed capacity.The physical machine quantity that existing IaaS cloud resource pool management technology is safeguarded is invariable, can not increase and decrease dynamically according to the variation of actual loading;
(2) be prone to the not enough and idle many situations of physical machine.Because the physical machine quantity in resource pool is constant, while easily there is high capacity without new physics machine can with and the too much situation of idle physical machine during low load;
(3) do not adopt controlling mechanism in advance.Existing cloud resource pool management strategy, does not carry out trend model and forecast to historical system service data, thereby management all has property afterwards opportunity, slower to the load acute variation reaction of burst.
Summary of the invention
The object of the invention is the problems referred to above that exist in order to solve prior art, optimize the efficiency of IaaS cloud resource management, the present invention proposes a kind of IaaS cloud scalable resource pool management method.
Technical scheme of the present invention is: a kind of IaaS cloud scalable resource pool management method, comprises following steps:
Step 1: obtain system information;
Step 2: analyze and decision-making, specifically comprise as follows step by step:
Step 21: data pre-service;
Step 22: calculate control decision reference value;
Step 23: send decision information;
Step 3: resource pool management;
Step 4: repeating step 1 is to step 3, until cloud application stops operation.
For the problems referred to above, the present invention also proposes a kind of IaaS cloud scalable resource pool management device, specifically comprises: analysis decision module and resource pool management module, wherein,
Analysis decision module comprises: for obtaining the system information acquiring unit of system information, for the pretreated data pretreatment unit of data, for calculating the controlled quentity controlled variable computing unit of control decision reference value, for sending the control decision unit of decision information.
For the problems referred to above, the present invention also proposes a kind of IaaS cloud scalable resource pool management server, specifically comprises: IaaS cloud scalable resource pool management device.
Beneficial effect of the present invention: method, device and the server of IaaS cloud scalable provided by the present invention resource pool management, the variation tendency of load and physical machine resource while being intended to follow the tracks of IaaS cloud system actual motion, then according to Future direction being arrived to the prediction of frequency, the operation whether dynamic decision expands/delete resource pool capacity, under guaranteeing as far as possible high capacity in system performance, the operating cost of system while reducing load, reach Optimization Balancing, take into account performance and the economy of cloud system.
Accompanying drawing explanation
Fig. 1 is schematic flow sheet of the present invention.
Fig. 2 is the composition structural representation of IaaS cloud scalable resource pool management device in the embodiment of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in further detail, but enforcement of the present invention and protection domain are not limited to this.
Technical matters to be solved by this invention is to provide method, device and the server of IaaS cloud scalable resource pool management.
Below the method, device and server are elaborated:
Fig. 1 is the method flow schematic diagram of the IaaS cloud scalable resource pool management in the embodiment of the present invention.The method comprises:
Step 1: obtain system information.
In this step, the information of obtaining comprises: the task quantity NT that epicycle is newly-increased
x; The task quantity WT that epicycle completes
x; The quantity KJ of the current physical machine in idle condition
x; The current physical machine quantity MJ in busy state
x; The quantity of moving in each busy state physical machine of task, T
i, 1≤i≤MJ
x.
Step 2: analyze and decision-making, specifically comprise as follows step by step:
Step 21: data pre-service.
Particularly, the average implementation rate ZXL of system information computing system obtaining according to step 1, the average task input of system rate SRL, epicycle can be accepted newly-increased number of tasks KJST
x:
The operation that mean{} is averaging for set, x is when front-wheel number, 1≤x< ∞, t is the interval time between predefined every wheel.Because 0<y<x, so WT
yfor certain quantity of finishing the work of taking turns before x.
The meaning directly perceived of above-mentioned formula is, if the task quantity that epicycle completes is not 0, and busy physical machine quantity is not 0, the inverse that the average task of each busy physical machine of take is processed quantity is system implementation rate, otherwise according to the average task handling rate of historical record is epicycle system implementation rate in the past.
The average task input of computing system rate SRL;
Here, because 0<y<x, so NT
yfor certain newly-increased task quantity of taking turns before x.
Epicycle can be accepted newly-increased number of tasks KJST
x;
Wherein, ZD represents the maximal value of the number of tasks that single physical machine can move simultaneously.
Epicycle can be accepted newly-increased number of tasks KJST
xmeaning directly perceived be, if the newly-increased task quantity of epicycle is less than or equal to the quantity of the current maximum admissible new tasks of system, using the newly-increased task quantity of epicycle as epicycle, newly-increased number of tasks can be accepted, otherwise take the quantity of the current maximum admissible new tasks of system, newly-increased number of tasks can be accepted as epicycle.
Calculating K JST
1to KJST
xthe rejecting of sequence the logarithm step-length average increment that disturbs of exceptional value:
loinc=mean{inc
u,v|0<u<v≤x,lq
u,v=0}
Wherein, lq
u,vfor judging whether it is the token variable of exceptional value of peeling off, inc
u,vthe logarithm step-length equivalence increment forming between u and v record value in expression sequence:
Lq
u,vobtaining value method be:
Wherein, a is parameter given in advance, between general desirable 0.1 to 0.5; Max{} asks maximum for set and operates; | inc
u,v| represent inc
u,vabsolute value, avg value is:
avg=mean{|inc
u,v||0<u<v≤x}
The meaning directly perceived of above-mentioned discriminant function is: the logarithm step-length equivalence quotient of difference sequence average logarithm step-length equivalence increment forming between a u and v record value is vast scale a also, and when all also large than the equivalent increment of single step logarithm step-lengths all between u and v record value, be considered as abnormal outlier.
Next calculate the newly-increased number of tasks FST of following expection:
Wherein, γ is that distance weakens the factor, and its effect is to make nearest historical record value larger on the impact of FST, and impact is more early less.
γ meets 0.5< γ <1, generally can be taken as 0.9.Part in formula after plus sige, for accepting being in history averaging of difference of newly-increased number of tasks and actual newly-increased number of tasks,
define the value of u, meaning be only calculate after ∑ on the occasion of situation, the ∑ of denominator part represents a plurality of 1 summation, is exactly in fact counter, has recorded and has met
the number of condition, so molecule denominator is divided by, equaling is exactly to ask to satisfy condition
every wheel in history can accept the average of newly-increased number of tasks.
Step 22: calculate control decision reference value.
The result obtaining according to step 21 is calculated the not enough probability BGY of following idle physical machine:
The visual interpretation of above-mentioned formula is that BGY is: system general assignment space, deduct the physical machine task space having taken, then the new task that deducts expection and increase takes up room, finally the not enough probability of remaining task space.
Step 23:
Send decision information, particularly, the not enough probability BGY of idle physical machine in future obtaining according to step 22 calculates decision information controlled quentity controlled variable KZL, and the computing method of this variable are:
Wherein, β
1and β
2for empirical value given in advance, meet 0< β
2<< β
1<0.5, generally can be by β
1be taken as 0.1, β
2be taken as 0.001, η
1and η
1for positive integer threshold value given in advance, between general desirable 5 to 10.
The meaning directly perceived of above-mentioned formula is: when the not enough probability of idle physical machine in future that step 22 calculates is greater than β
1, and the difference that the newly-increased task of epicycle deducts the quantity of finishing the work is greater than η
1, controlled quentity controlled variable is set as to 1; Come the not enough probability of idle physical machine to be less than β
2, and the epicycle difference that deducts newly-increased task quantity of finishing the work is greater than η
2, controlled quentity controlled variable is set as to-1; Other situations, setup control amount is 0.
Step 3: resource pool management.
If KZL is 0, keep the physical machine quantity in existing resource pool constant; If KZL is 1, calls a new idle physical machine and enter resource pool; If KZL is-1, and currently there is at least one idle physical machine, close at random an idle physical machine; If KZL is-1, and in current resource pool, all physical machine are all busy, keep the physical machine quantity in existing resource pool constant.
Step 4: wait for that t, after the time, repeats above-mentioned steps, until IaaS cloud system is out of service.
In order to solve technical matters to be solved by this invention, the present embodiment also provides a kind of IaaS cloud scalable resource pool management device, specifically comprises: analysis decision module and resource pool management module, and wherein, analysis decision module comprises:
For obtaining the system information acquiring unit of system information;
The information of obtaining according to described system information acquiring unit is carried out the pretreated data pretreatment unit of data;
The data that calculate according to described data pretreatment unit are calculated the controlled quentity controlled variable computing unit of control decision reference value;
The data that obtain according to described controlled quentity controlled variable computing unit are sent the control decision unit of decision information.
The information that the system information acquiring unit here obtains comprises: the task quantity NT that epicycle is newly-increased
x; The task quantity WT that epicycle completes
x; The quantity KJ of the current physical machine in idle condition
x; The current physical machine quantity in busy state, MJ
x; The quantity T of moving in each busy state physical machine of task
i, 1≤i≤MJ
x;
The system information acquiring unit here sends to described data pretreatment unit after obtaining above-mentioned information, and described data pretreatment unit can be accepted newly-increased number of tasks KJST according to the average implementation rate ZXL of information computing system, the average task input of the system rate SRL, the epicycle that receive
x,
Wherein, average implementation rate ZXL is:
The operation that mean{} is averaging for set, x is when front-wheel number, 1≤x< ∞, t is the interval time between predefined every wheel.
Average task input rate SRL is:
NT
xthe task quantity that epicycle is newly-increased, because 0<y<x, so NT
yfor certain newly-increased task quantity of taking turns before x.
Epicycle can be accepted newly-increased number of tasks KJST
xfor:
Wherein, ZD represents the maximal value of the number of tasks that single physical machine can move simultaneously.
The data pretreatment unit here also can be accepted newly-increased number of tasks KJST according to the average implementation rate ZXL of system, the average task input of system rate SRL and epicycle
x, calculating K JST
1to KJST
xsequence has been rejected the logarithm step-length average increment that exceptional value is disturbed:
loinc=mean{inc
u,v|0<u<v≤x,lq
u,v=0}
Wherein, lq
u,vfor judging whether it is the token variable of exceptional value of peeling off, inc
u,vthe logarithm step-length equivalence increment forming between u and v record value in expression sequence:
Lq
u,vobtaining value method be:
Wherein, a is parameter given in advance, between general desirable 0.1 to 0.5; Max{} asks maximum for set and operates; | inc
u,v| represent inc
u,vabsolute value, avg value is:
avg=mean{|inc
u,v||0<u<v≤x}
The data pretreatment unit here also calculates the newly-increased number of tasks FST of following expection according to logarithm step-length average increment loinc:
Wherein, γ is that predefined distance weakens the factor.
The result that the described controlled quentity controlled variable computing unit here obtains according to described data pretreatment unit is calculated the not enough probability BGY of following idle physical machine:
Described not enough probability BGY is: system general assignment space, deduct the physical machine task space having taken, then the new task that deducts expection and increase takes up room, finally the not enough probability of remaining task space.
The not enough probability BGY of idle physical machine in future that the control decision unit here obtains according to described controlled quentity controlled variable computing unit calculates decision information controlled quentity controlled variable KZL, and the computation process of this variable is:
Wherein, β
1and β
2for threshold value given in advance.
The resource pool management module here comprises resource pool management unit, and the decision information controlled quentity controlled variable KZL that described resource pool management unit obtains according to described control decision unit manages as follows:
If KZL is 0, keep the physical machine quantity in existing resource pool constant; If KZL is 1, calls a new idle physical machine and enter resource pool; If KZL is-1, and currently there is at least one idle physical machine, close at random an idle physical machine; If KZL is-1, and in current resource pool, all physical machine are all busy, keep the physical machine quantity in existing resource pool constant.
In order to solve technical matters to be solved by this invention, the present embodiment also provides a kind of IaaS cloud scalable resource pool management server, specifically comprises: IaaS cloud scalable resource pool management device.The IaaS cloud scalable resource pool management device that the embodiment of the present invention provides, can be deployed in an existing server, also can dispose with in a server that is exclusively used in the resource pool management of IaaS cloud scalable arranging separately.For this reason, the invention provides a kind of server, comprise the IaaS cloud scalable resource pool management device that the embodiment of the present invention provides, the composition structural representation of this IaaS cloud scalable resource pool management device as shown in Figure 2.One of ordinary skill in the art will appreciate that the process that realizes IaaS cloud scalable resource pool management in above-described embodiment method, can complete by the relevant hardware of programmed instruction, described program can be stored in the readable storage medium storing program for executing of IaaS cloud scalable resource pool management device, and this program is carried out the corresponding step in said method when carrying out.Described storage medium can be as: ROM/RAM, magnetic disc, CD etc.
With respect to traditional IaaS cloud resource pool management method, method of the present invention, transposition and system have the following advantages: carry out dynamically decision-making calculating, determine whether and need to expand or delete the physical machine quantity in resource pool, and then can determine dynamically the scale of resource pool, take into account performance and operating cost; By historic task executing data and task, arrive data, the to-be of anticipation cloud system, makes control decision targetedly, makes system make response more timely to the sharply variation of load, reaches real-time corresponding task load variations; When processing historical data, rejected the impact of the abnormal data that wherein peels off, make control decision more accurate.
Claims (10)
1. an IaaS cloud scalable resource pool management method, comprises following steps:
Step 1: obtain system information;
Step 2: analyze and decision-making, specifically comprise as follows step by step:
Step 21: data pre-service;
Step 22: calculate control decision reference value;
Step 23: send decision information;
Step 3: resource pool management;
Step 4: repeating step 1 is to step 3, until cloud application stops operation.
2. IaaS cloud scalable resource pool management method according to claim 1, is characterized in that, the information that step 1 is obtained comprises: the task quantity NT that epicycle is newly-increased
x; The task quantity WT that epicycle completes
x; The quantity KJ of the current physical machine in idle condition
x; The current physical machine quantity MJ in busy state
x; The quantity of moving in each busy state physical machine of task, T
i, 1≤i≤MJ
x.
3. IaaS cloud scalable resource pool management method according to claim 1, is characterized in that, the pretreated detailed process of data described in step 2 is as follows:
The average implementation rate ZXL of system information computing system obtaining according to step 1, the average task input of system rate SRL, epicycle can be accepted newly-increased number of tasks KJST
x,
Wherein, average implementation rate ZXL is:
The operation that mean{} is averaging for set, x is when front-wheel number, 1≤x< ∞, t is the interval time between predefined every wheel;
Average task input rate SRL is:
Epicycle can be accepted newly-increased number of tasks KJST
xfor:
Wherein, ZD represents the maximal value of the number of tasks that single physical machine can move simultaneously;
Calculating K JST
1to KJST
xthe rejecting of sequence the logarithm step-length average increment that disturbs of exceptional value:
loinc=mean{inc
u,v|0<u<v≤x,lq
u,v=0}
Wherein, lq
u,vfor judging whether it is the token variable of exceptional value of peeling off, inc
u,vthe logarithm step-length equivalence increment forming between u and v record value in expression sequence:
Lq
u,vobtaining value method be:
Wherein, a is parameter given in advance, and max{} asks maximum for set and operates; | inc
u,v| represent inc
u,vabsolute value, avg value is:
avg=mean{|inc
u,v||0<u<v≤x}
Calculate the newly-increased number of tasks FST of following expection:
Wherein, γ is that predefined distance weakens the factor.
4. IaaS cloud scalable resource pool management method according to claim 3, is characterized in that, the process of the calculating control decision reference value described in step 2 is as follows:
The result obtaining according to step 21 is calculated the not enough probability BGY of following idle physical machine:
Described not enough probability BGY is: system general assignment space, deduct the physical machine task space having taken, then the new task that deducts expection and increase takes up room, finally the not enough probability of remaining task space.
5. IaaS cloud scalable resource pool management method according to claim 4, is characterized in that, described in step 2 to send decision information detailed process as follows:
The not enough probability BGY of idle physical machine in future obtaining according to step 22 calculates decision information controlled quentity controlled variable KZL, and the computing method of this variable are:
Wherein, β
1and β
2for threshold value given in advance.
6. IaaS cloud scalable resource pool management method according to claim 5, is characterized in that, the resource pool management detailed process described in step 3 is as follows:
If KZL is 0, keep the physical machine quantity in existing resource pool constant; If KZL is 1, calls a new idle physical machine and enter resource pool; If KZL is-1, and currently there is at least one idle physical machine, close at random an idle physical machine; If KZL is-1, and in current resource pool, all physical machine are all busy, keep the physical machine quantity in existing resource pool constant.
7. an IaaS cloud scalable resource pool management device, specifically comprises: analysis decision module and resource pool management module,
Wherein, analysis decision module comprises:
For obtaining the system information acquiring unit of system information;
The information of obtaining according to described system information acquiring unit is carried out the pretreated data pretreatment unit of data;
The data that calculate according to described data pretreatment unit are calculated the controlled quentity controlled variable computing unit of control decision reference value;
The data that obtain according to described controlled quentity controlled variable computing unit are sent the control decision unit of decision information.
8. IaaS cloud scalable resource pool management device according to claim 7, is characterized in that,
The information that described system information acquiring unit obtains comprises: the task quantity that epicycle is newly-increased, NT
x; The task quantity that epicycle completes, WT
x; The quantity of the current physical machine in idle condition, KJ
x; The current physical machine quantity in busy state, MJ
x; The quantity of moving in each busy state physical machine of task, T
i, 1≤i≤MJ
x;
Described system information acquiring unit sends to described data pretreatment unit after obtaining above-mentioned information, and described data pretreatment unit can be accepted newly-increased number of tasks KJST according to the average implementation rate ZXL of information computing system, the average task input of the system rate SRL, the epicycle that receive
x,
Wherein, average implementation rate ZXL is:
The operation that mean{} is averaging for set, x is when front-wheel number, 1≤x< ∞, t is the interval time between predefined every wheel.
Average task input rate SRL is:
Epicycle can be accepted newly-increased number of tasks KJST
xfor:
Wherein, ZD represents the maximal value of the number of tasks that single physical machine can move simultaneously;
Described data pretreatment unit also can be accepted newly-increased number of tasks KJST according to the average implementation rate ZXL of system, the average task input of system rate SRL and epicycle
x, calculating K JST
1to KJST
xsequence has been rejected the logarithm step-length average increment that exceptional value is disturbed:
loinc=mean{inc
u,v|0<u<v≤x,lq
u,v=0}
Wherein, lq
u,vfor judging whether it is the token variable of exceptional value of peeling off, inc
u,vthe logarithm step-length equivalence increment forming between u and v record value in expression sequence:
Lq
u,vobtaining value method be:
Wherein, a is parameter given in advance, between general desirable 0.1 to 0.5; Max{} asks maximum for set and operates; | inc
u,v| represent inc
u,vabsolute value, avg value is:
avg=mean{|inc
u,v||0<u<v≤x}
Described data pretreatment unit also calculates the newly-increased number of tasks FST of following expection according to logarithm step-length average increment loinc:
Wherein, γ is that predefined distance weakens the factor.
9. IaaS cloud scalable resource pool management device according to claim 7, is characterized in that, the result that described controlled quentity controlled variable computing unit obtains according to described data pretreatment unit is calculated the not enough probability BGY of following idle physical machine:
Described not enough probability BGY is: system general assignment space, deduct the physical machine task space having taken, then the new task that deducts expection and increase takes up room, finally the not enough probability of remaining task space;
The not enough probability BGY of idle physical machine in future that described control decision unit obtains according to described controlled quentity controlled variable computing unit calculates decision information controlled quentity controlled variable KZL, and the computation process of this variable is:
Wherein, β
1and β
2for threshold value given in advance;
Described resource pool management module comprises resource pool management unit, and the decision information controlled quentity controlled variable KZL that described resource pool management unit obtains according to described control decision unit manages as follows:
If KZL is 0, keep the physical machine quantity in existing resource pool constant; If KZL is 1, calls a new idle physical machine and enter resource pool; If KZL is-1, and currently there is at least one idle physical machine, close at random an idle physical machine; If KZL is-1, and in current resource pool, all physical machine are all busy, keep the physical machine quantity in existing resource pool constant.
10. an IaaS cloud scalable resource pool management server, specifically comprises the IaaS cloud scalable resource pool management device described in claim 7 to 9 any one claim.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410188323.8A CN104063282B (en) | 2014-05-06 | 2014-05-06 | Management method, device and server for IaaS cloud variable scale resource pool |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410188323.8A CN104063282B (en) | 2014-05-06 | 2014-05-06 | Management method, device and server for IaaS cloud variable scale resource pool |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104063282A true CN104063282A (en) | 2014-09-24 |
CN104063282B CN104063282B (en) | 2017-02-15 |
Family
ID=51551005
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410188323.8A Active CN104063282B (en) | 2014-05-06 | 2014-05-06 | Management method, device and server for IaaS cloud variable scale resource pool |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104063282B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104503846A (en) * | 2015-01-22 | 2015-04-08 | 成都派沃特科技有限公司 | Resource management system based on cloud computing system |
CN104765435A (en) * | 2015-03-31 | 2015-07-08 | 重庆大学 | Method, device and system for delaying starting up of cloud data center host computer |
CN106201847A (en) * | 2016-06-30 | 2016-12-07 | 重庆大学 | Consider method for allocating tasks, device and the system of the decay of cloud platform host performance |
CN108270833A (en) * | 2016-12-31 | 2018-07-10 | ***通信集团安徽有限公司 | Render automatic scheduling method, the apparatus and system of cloud resource |
CN114157673A (en) * | 2021-11-30 | 2022-03-08 | 北京知道创宇信息技术股份有限公司 | CDN system node management method and device, storage medium and electronic equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103365713A (en) * | 2012-04-01 | 2013-10-23 | 华为技术有限公司 | Resource dispatch and management method and device |
CN103576829A (en) * | 2012-08-01 | 2014-02-12 | 复旦大学 | Hybrid genetic algorithm based dynamic cloud-computing virtual machine scheduling method |
US20140095721A1 (en) * | 2012-06-27 | 2014-04-03 | International Business Machines Corporation | Instantiating resources of an it-service |
-
2014
- 2014-05-06 CN CN201410188323.8A patent/CN104063282B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103365713A (en) * | 2012-04-01 | 2013-10-23 | 华为技术有限公司 | Resource dispatch and management method and device |
US20140095721A1 (en) * | 2012-06-27 | 2014-04-03 | International Business Machines Corporation | Instantiating resources of an it-service |
CN103576829A (en) * | 2012-08-01 | 2014-02-12 | 复旦大学 | Hybrid genetic algorithm based dynamic cloud-computing virtual machine scheduling method |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104503846A (en) * | 2015-01-22 | 2015-04-08 | 成都派沃特科技有限公司 | Resource management system based on cloud computing system |
CN104503846B (en) * | 2015-01-22 | 2018-01-30 | 成都派沃特科技有限公司 | A kind of resource management system based on cloud computing system |
CN104765435A (en) * | 2015-03-31 | 2015-07-08 | 重庆大学 | Method, device and system for delaying starting up of cloud data center host computer |
CN104765435B (en) * | 2015-03-31 | 2018-01-16 | 重庆大学 | A kind of methods, devices and systems of cloud data center main frame delay start |
CN106201847A (en) * | 2016-06-30 | 2016-12-07 | 重庆大学 | Consider method for allocating tasks, device and the system of the decay of cloud platform host performance |
CN106201847B (en) * | 2016-06-30 | 2019-04-23 | 重庆大学 | Consider method for allocating tasks, the device and system of the decaying of cloud platform host performance |
CN108270833A (en) * | 2016-12-31 | 2018-07-10 | ***通信集团安徽有限公司 | Render automatic scheduling method, the apparatus and system of cloud resource |
CN108270833B (en) * | 2016-12-31 | 2021-07-16 | ***通信集团安徽有限公司 | Automatic scheduling method, device and system for rendering cloud resources |
CN114157673A (en) * | 2021-11-30 | 2022-03-08 | 北京知道创宇信息技术股份有限公司 | CDN system node management method and device, storage medium and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN104063282B (en) | 2017-02-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102063336B (en) | Distributed computing multiple application function asynchronous concurrent scheduling method | |
CN104991830B (en) | YARN resource allocations and energy-saving scheduling method and system based on service-level agreement | |
CN102156665B (en) | Differential serving method for virtual system competition resources | |
CN104063282A (en) | Management method, device and server for IaaS cloud variable scale resource pool | |
CN102799512B (en) | Virtual machine monitoring method in vector-autoregression-based cloud computing | |
CN106773711B (en) | A kind of the hybrid tasks scheduling method and model of railway locomotive operation steerable system | |
CN104657220A (en) | Model and method for scheduling for mixed cloud based on deadline and cost constraints | |
CN102521055B (en) | Virtual machine resource allocating method and virtual machine resource allocating system | |
CN103745225A (en) | Method and system for training distributed CTR (Click To Rate) prediction model | |
CN102609303B (en) | Slow-task dispatching method and slow-task dispatching device of Map Reduce system | |
Bi et al. | SLA-based optimisation of virtualised resource for multi-tier web applications in cloud data centres | |
CN105893148A (en) | Low-energy-consumption aporadic task scheduling method based on RM strategy | |
CN102508717B (en) | Memory scheduling method and memory scheduling device for multi-core processor | |
CN112685153A (en) | Micro-service scheduling method and device and electronic equipment | |
CN104111875B (en) | Cloud data center increases number of tasks device for controlling dynamically, system and method newly | |
CN107479944A (en) | Mix the adaptive thermophoresis dispatching method of virutal machine memory and system under cloud mode | |
CN113515382A (en) | Cloud resource allocation method and device, electronic equipment and program product | |
US9983908B1 (en) | Adjusting allocation of selected resources for capped and uncapped virtual machines | |
CN103442087B (en) | A kind of Web service system visit capacity based on response time trend analysis controls apparatus and method | |
CN106027318A (en) | Cloud computing-based two-level optimal scheduling management platform for virtual machine | |
Kumbhare et al. | Dynamic power management for value-oriented schedulers in power-constrained HPC system | |
CN103685541B (en) | IaaS cloud system operating rate device for controlling dynamically, system and method | |
CN114138453B (en) | Resource optimization allocation method and system suitable for edge computing environment | |
CN116028193A (en) | Big data task dynamic high-energy-efficiency scheduling method and system for mixed part cluster | |
CN110069319A (en) | A kind of multiple target dispatching method of virtual machine and system towards cloudlet resource management |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |