CN102185759A - Multi-physical server load equalizing method and device capable of meeting requirement characteristic - Google Patents
Multi-physical server load equalizing method and device capable of meeting requirement characteristic Download PDFInfo
- Publication number
- CN102185759A CN102185759A CN2011100903353A CN201110090335A CN102185759A CN 102185759 A CN102185759 A CN 102185759A CN 2011100903353 A CN2011100903353 A CN 2011100903353A CN 201110090335 A CN201110090335 A CN 201110090335A CN 102185759 A CN102185759 A CN 102185759A
- Authority
- CN
- China
- Prior art keywords
- physical server
- utilance
- physical
- virtual machine
- cpu
- 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 embodiment of the invention discloses a multi-physical server load equalizing method and device. The method comprises the following steps: virtual servers are dispensed to suitable physical servers when the load of each physical server is not equalizing and the virtual servers to be dispensed exist, thus well realizing the equalizing load among various physical servers; the load equalizing method and device simultaneously consider the equalizing load among the data center physical servers and the equalizing load among the three properties (central processing unit, internal memory and network wideband) of each physical server, and meet the different requirement characteristics of the data center physical servers, thus ensuring each physical server to work normally and provide high-performance service. The multi-physical server load equalizing method and device are applicable to load equalization on a plurality of physical servers of the cloud calculation data center.
Description
1. technical field
The present invention relates to the computer communication technology field, particularly a kind of method and device of realizing many physical servers load balancing.
2. background technology
The present invention relates to the dynamic load equilibrium technology of a plurality of physical servers of cloud computing data center.There is a large amount of inconsistent physical servers of specification configuration possibility in cloud data center, and by Intel Virtualization Technology, each physical server can fictionalize a plurality of virtual machines.Cloud computing data center dynamically distributes to automation the user with these virtual machines by user's request specification (may be inconsistent).But because the configuration of the specification of user's requirement specification and all physical servers of data center is inconsistent, if adopt simple allocation schedule method, wheel for example commonly used changes method, the weighted round robin method, minimum load (or link number) is preferential, weighting minimum load precedence method, Hash method etc. are difficult to reach the physical server load balancing.Thereby can cause unbalanced and other relevant issues of service performance.
What a turnaround time (for example physical server number) is wheel commentaries on classics method (Round Robin) normally preestablish, and successively the virtual machine of user's request distributed to different physical servers, restarts a new wheel after finish a turnaround time and changes.Wheel commentaries on classics method can not solve physical server and the user's request unbalanced problem of the load that causes that causes diversified in specifications.The weighted round robin method is set weights to physical server in advance, distribute in the process of virtual machine in load balancing, wheel changes selects physical server, if the weights of selecteed physical server are 0, then skip this server and select next, weights as selecteed server are not 0, then choose this server also the weights of this server to be subtracted 1, and wheel changes on the basis of last time selecting in follow-up being chosen in.Be respectively 1,2 with weights, three physical server (PM1 of 3, PM2 PM3) is example, selects first physical server PM1 for the first time, its weights are kept to 0, select second physical server PM2 for the second time, and its weights are kept to 1, select the 3rd physical server PM3 for the third time, its weights are kept to 2, the four next rounds and forward the first station server PM1 to, but its weights are 0, continue wheel and change, and select the second station server PM2, its weights are kept to 0 simultaneously ...The order of selecting for six times is successively: PM1, PM2, PM3, PM2, PM3, PM3.The service times that weights are high like this server obtains just is directly proportional with its weights, still still has the unbalanced problem of load when causing when user's request is diversified in specifications.The weighted round robin method need be revised the weights of each station server in balancing procedure in addition, and these public variables need carry out locking and de-locking, influence execution speed.Minimum load (or link number) precedence method is when distributing virtual machine at every turn, at first checks the loading condition of interior all physical servers for the previous period, and always selecting for the previous period, the physical server of internal burden minimum distributes virtual machine.Two figure of Fig. 1, Fig. 2 can describe this technical scheme implementation process.Task one is scheduled on the website 2 of visit load minimum ..., task four is assigned to website 3.This method can't solve user's request specification and the unbalanced problem of the physical server specification inconsistent load that causes of configuration equally.Weighting minimum load precedence method, be similar to the weighted round robin method, dispose inconsistent physical server for specification and set different weights, consider weights when selecting to distribute virtual machine, this method can't solve the unbalanced problem of the sharp physical server specification inconsistent load that causes of configuration of user's request specification equally.Weighted round robin method and weighting minimum load method execution speed are subjected to certain influence because of need calculate weights.The Hash method designs a hash function in advance, be used to shine upon the virtual machine of user's request to corresponding physical server, execution speed is very fast, has not only satisfied the user specification demand but also satisfied the configuration of physical server specification inconsistent and to solve the unbalanced problem of load almost be impossible but how to design a hash function.
In realizing process of the present invention, the inventor finds that there are the following problems at least in the prior art: above-mentioned simple allocation schedule method all can not solve the inconsistent unbalanced problem of each physical server load that causes of user specification demand and physical server specification configuration.
Therefore the present invention designs load-balancing algorithm and the device that a foundation is dynamically adjusted, and solves the inconsistent unbalanced problem of each physical server load that causes of user specification demand and physical server specification configuration better.
3. summary of the invention
Embodiments of the invention provide a kind of method and device of realizing many physical servers load balancing, can realize the load balancing of many physical servers well.
The technical scheme that the embodiment of the invention adopts is:
A kind of method that realizes many physical servers load balancing comprises:
The algorithm that distributes: at first physical server is arranged by the utilance ascending order of the type according to the virtual machine type (cpu, internal memory, network) of user's request.Utilance according to the type is divided into a plurality of intervals with physical server then, and the size in each interval can dynamically arrange (such as being an interval with 0.05), finds out all physical servers in the interval at the minimum physical server place of utilance then.The virtual machine examination is distributed on all physical servers in this interval, the cpu utilance of each physical server behind the difference Computational Physics server-assignment virtual machine, memory usage, network utilization three's variance obtains the load balancing value of each physical server.Concrete computing formula is as follows:
CombinedBalanceLevel＝(AVG-CPU_utility)*(AVG-CPU_utility)+(AVG-MEM_utility)*(AVG-MEM_utility)+(AVG-BW_utility)*(AVG-BW_utility) (1)
AVG＝(CPU_utility+MEM_utility+BW_utility)/3 (2)
CPU_utility wherein, MEM_utility, BW_utility represent physical server cpu utilance, memory usage, network bandwidth utilance respectively.AVG represents they three's average.
Choose the minimum physical server of load balancing value and begin to distribute virtual machine, as long as be no more than the heap(ed) capacity of this physical server behind this virtual machine of distribution, then become the distribution of work, otherwise getting next load balancing value time little physical server distributes, if should all can not distribute by all physical servers of interval, then take out next interval physical server and distribute, till can distributing.
Stand-by period prediction: if all physical servers all can not distribute, then this task is added waiting list, then according to task in the waiting list of front up to operation time of required wait and the time that the number of tasks in the waiting list is predicted the wait of this task, the time of notifying the user to wait for simultaneously.
Migration algorithm: according to the cpu of our setting, internal memory, the maximum disparity of the network bandwidth, successively relatively in the current scheduling territory difference between peak use rate and minimum utilance whether surpass this value, move if surpass then need begin, the utilance of that all physical servers are surpassed sorts, find the physical server of utilance maximum, and then find the virtual machine of the type utilance minimum from physical machine, to take out from this physical server, call allocation algorithm then and redistribute this virtual machine.After distributing once, continue the circulation migration, up to the requirement less than maximum disparity, perhaps occurred repeatedly till the situation of migration, then this time migration finishes.
A kind of device of realizing many physical servers load balancing comprises:
Distribution module, be used for distributing current newly to the virtual machine task requests or the virtual machine task requests of waiting list.
The stand-by period prediction module is used to predict the stand-by period when the virtual machine requested resource is not enough.
Transferring module is used for when the difference between each physical server optimum rate of utilization and the minimum utilance during greater than preset difference value, and the virtual machine on the physical server that utilance is the highest moves, up between the two difference less than preset difference value.
The method and the device of load balancing between the many physical servers of the realization that the embodiment of the invention provides, at first according to the type (cpu of virtual machine, internal memory, network), physical server utilance with corresponding types sorts then, find out the interval of corresponding utilance minimum, calculate this interval load balancing value, virtual machine is distributed on the physical machine of load balancing value minimum.When Task Distribution is failed, then enter the stand-by period prediction module, calculate the stand-by period and notify the user, so that the user makes corresponding selection.During greater than preset difference value, the virtual machine on the physical server that utilance is the highest moves, till difference is less than preset difference value in the difference of physical server optimum rate of utilization and minimum utilance.Compared with prior art, the embodiment of the invention can be between each physical server load unbalanced and when having virtual server to be allocated, virtual server is assigned on the suitable physical server, thereby can realize the load balancing between many physical servers well, this load balancing has been considered three kinds of attribute (cpu of load balancing between data center's physical server and each physical server simultaneously, internal memory, the network bandwidth) load balancing between guarantees that each physical server can normally make and provide high performance service down.
4. description of drawings
Fig. 1, Fig. 2 are minimum load priority algorithm example;
The method flow diagram that Fig. 3 provides for the embodiment of the invention one;
The method flow diagram that Fig. 4 provides for the embodiment of the invention two;
Fig. 5 is existing dispatching patcher reference architecture schematic diagram;
Fig. 6 has marked main functional modules;
By with description of drawings (corresponding text all describes in detail), it is easier to understand that feature of the present invention will become.
5. embodiment
The embodiment of the invention provides a kind of device (shown in figure-6) of realizing data center's balancing resource load, comprising:
Comprise:
Select control module 41, be used to obtain the current utilance of each physical machine attribute in the dispatching zone, and determine that according to described current utilance surplus capacity satisfies the physical machine of current allocating task, described attribute comprises cpu load, internal memory load and offered load;
Distribute Executive Module 43, be used to select the physical machine of described load balancing value minimum to distribute described current allocating task.
For the advantage that makes technical solution of the present invention is clearer, the present invention is elaborated below in conjunction with drawings and Examples.
Embodiment one
Present embodiment provides a kind of method that realizes load balancing between many physical servers, and as shown in Figure 3, described method comprises:
101, at first physical server is arranged by the utilance ascending order of the type according to the virtual machine type (cpu, internal memory, network) of user's request.
102, according to the utilance of the type physical server is divided into a plurality of intervals then, the size in each interval can dynamically arrange (such as being an interval with 0.05), finds out all physical servers in the interval at the minimum physical server place of utilance then.
103, the virtual machine examination is distributed on all physical servers in this interval, the cpu utilance of each physical server behind the difference Computational Physics server-assignment virtual machine, memory usage, network utilization three's variance obtains the load balancing value of each physical server.
104, choose the minimum physical server of load balancing value and begin to distribute virtual machine,, then become the distribution of work as long as be no more than the heap(ed) capacity of this physical server behind this virtual machine of distribution.
105 otherwise get next load balancing value time little physical server and distribute, if should all can not distribute by all physical servers of interval, then take out next interval physical server and distribute, till can distributing.
Concrete virtual machine is assigned on the suitable physical server can be with reference to following process:
Table-1 initial situation
New task #1 arrives cpu=2 internal memory=1 network bandwidth=2 types: the CPU type
The interval of cpu utilance minimum is 0.00-0.05, and this interval physical server has PM1, PM2, PM3.
Table-2 computational load equilibrium values
Wherein the every value of PM1 is calculated as follows (get decimal point after 4):
CPU_utility＝2/6.4＝0.3125
Mem_utility＝1/12.0＝0.0833
BW_utility＝2/20.0＝0.1
Load balancing value=[0.3125-(0.3125+0.0833+0.1)/3] 2+[0.0833-(0.3125+0.0833+0.1)/3] 2+[0.1-(0.3125+0.0833+0.1)/3] other compute classes of 2=0.0326 are seemingly.
Table-3 new task #1 distribute the every numerical value in back
New task #2 arrives cpu=2 internal memory=2 network bandwidths=4 types: the CPU type
The interval of cpu utilance minimum is 0.00-0.05, and this interval physical server has PM2, PM3.
Table-4 computational load equilibrium values
After table-5 new task #2 distribute
New task #3 arrives cpu=2 internal memory=2 network bandwidths=8 types: network bandwidth type
The interval of network bandwidth utilance minimum is 0.00-0.05, and this interval physical server has PM3.
Table-6 new task #3 arrive back computational load equilibrium value
After table-7 new task #3 distribute
New task #4 arrives cpu=1 internal memory=2 network bandwidths=4 types: the internal memory type
The interval of memory usage minimum is 0.05-0.10, and this interval physical server has PM1, PM3.
The computational load equilibrium value
Table-8 new task #4 arrive back computational load equilibrium value
After table-9 new task #4 distribute
Embodiment two
Present embodiment provides a kind of method that realizes load balancing between many physical servers, and as shown in Figure 4, described method comprises:
201, the total size of the cpu that obtains all physical servers of current time, the cpu utilance, internal memory is size always, memory usage, network is size always, network utilization.
202, accept new task requests (definition priority four formations from high to low are respectively wait, request, tuning, deletion formation), the tuning task adds the tuning formation, the formation (by the priority ordering of time started) that joins request of domestic consumer's task.
203, check waiting list, if wait is not empty, the task of taking out in the waiting list is distributed, and concrete assigning process is referring to 208,209,210,211.
204, check request queue, if request queue is not empty, and the time started of task arrive, then begin to distribute.If become the distribution of work then this request added the deletion formation.
205, check the tuning formation, if tuning team is not empty, then begin tuning, the tuning mode is referring to 212,213.
206, check the deletion formation, time is up to see if there is the deletion of task.
Time is up if 207 have the task deletion, then deletes task.
208, according to the virtual machine type (cpu, internal memory, network) of user's request physical server is arranged by the utilance ascending order of the type.
209, according to the utilance of the type physical server is divided into a plurality of intervals then, the size in each interval can dynamically arrange (such as being an interval with 0.05), finds out all physical servers in the interval at the minimum physical server place of utilance then.
210, this virtual machine examination is distributed on all physical servers in this interval, the cpu utilance of each physical server behind the difference Computational Physics server-assignment virtual machine, memory usage, network utilization three's variance obtains the load balancing value of each physical server.
211, choose the minimum physical server of load balancing value and begin to distribute virtual machine, as long as be no more than the heap(ed) capacity of this physical server behind this virtual machine of distribution, then become the distribution of work, otherwise getting next load balancing value time little physical server distributes, if should all can not distribute by all physical servers of interval, then take out next interval physical server and distribute, till can distributing.If all physical servers all can not distribute, then this task is added waiting list, then according to task in the waiting list of front up to operation time of required wait and the time that the number of tasks in the waiting list is predicted the wait of this task, the time of notifying the user to wait for simultaneously.
212, the cpu that sets according to us, internal memory, the maximum disparity of the network bandwidth, successively relatively in the current scheduling territory difference between peak use rate and minimum utilance whether surpass this value.
If 213 surpass then need begin migration, the utilance of that all physical servers are surpassed sorts, find the physical server of utilance maximum, and then find the virtual machine of the type utilance minimum from physical machine, to take out from this physical server, call allocation algorithm then and redistribute this virtual machine.After distributing once, continue the circulation migration, up to the requirement less than maximum disparity, perhaps occurred repeatedly till the situation of migration, then this time migration finishes.
One of ordinary skill in the art will appreciate that all or part of flow process that realizes in the foregoing description method, be to finish with relevant hardware by computer program instructions, described program can be stored in the computer read/write memory medium, this program can comprise the flow process as the embodiment of above-mentioned each side method when carrying out.Wherein, described storage medium can be magnetic disc, CD, read-only storage memory body (Read-Only Memory, ROM) or at random store memory body (Random Access Memory, RAM) etc.
The above; only be the specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of claim.
Claims (11)
1. the algorithm of Fen Peiing: at first physical server is arranged by the utilance ascending order of the type according to the virtual machine type (cpu, internal memory, network) of user's request.Utilance according to the type is divided into a plurality of intervals with physical server then, and the size in each interval can dynamically arrange (such as being an interval with 0.05), finds out all physical servers in the interval at the minimum physical server place of utilance then.The virtual machine examination is distributed on all physical servers in this interval, the cpu utilance of each physical server behind the difference Computational Physics server-assignment virtual machine, memory usage, network utilization three's variance obtains the load balancing value of each physical server.Choose the minimum physical server of load balancing value and begin to distribute virtual machine, as long as be no more than the heap(ed) capacity of this physical server behind this virtual machine of distribution, then become the distribution of work, otherwise getting next load balancing value time little physical server distributes, if should all can not distribute by all physical servers of interval, then take out next interval physical server and distribute, till can distributing.
2. by stages method: physical server is divided into a plurality of intervals, the size in each interval can dynamically arrange, all between the physical server location of utilance minimum, distribute at every turn, can guarantee like this to assign the task to the little physical server of utilance earlier, guarantee that whole data center is relatively more balanced.Each simultaneously physical server of only seeking from an interval has reduced running time of algorithm greatly.
3. the computing formula of load balancing value:
CombinedBalanceLevel＝(AVG-CPU_utility)*(AVG-CPU_utility)+(AVG-MEM_utility)*(AVG-MEM_utility)+(AVG-BW_utility)*(AVG-BW_utility) (1)
AVG＝(CPU_utility+MEM_utility+BW_utility)/3 (2)
CPU_utility wherein, MEM_utility, BW_utility represent physical server cpu utilance, memory usage, network bandwidth utilance respectively.AVG represents they three's average.Each physical server of load balancing value minimum of selecting distributes the cpu that just can guarantee single physical server self, and internal memory, network bandwidth attribute are relatively more balanced.
4. the distribution method of the characteristic that satisfies the demands: physical server is sorted when selecting suitable physical server all is to sort according to the virtual machine type of asking at every turn, when the virtual machine type is the cpu type, physical machine is then pressed the ordering of cpu utilance, and in like manner the internal memory and the network bandwidth also are.The different demands that can guarantee the user like this can both well be met.The physical server of data center can obtain the most effective utilization simultaneously.
5. stand-by period Forecasting Methodology: if all physical servers all can not distribute, then this task is added waiting list, then according to task in the waiting list of front up to operation time of required wait and the time that the number of tasks in the waiting list is predicted the wait of this task, the time of notifying the user to wait for simultaneously.
6. migration algorithm: the cpu that sets according to us, internal memory, the maximum disparity of the network bandwidth, successively relatively in the current scheduling territory difference between peak use rate and minimum utilance whether surpass this value, move if surpass then need begin, the utilance of that all physical servers are surpassed sorts, find the physical server of utilance maximum, and then find the virtual machine of the type utilance minimum from physical machine, to take out from this physical server, call allocation algorithm then and redistribute this virtual machine.After distributing once, continue the circulation migration, up to the requirement less than maximum disparity, perhaps occurred repeatedly till the situation of migration, then this time migration finishes.
7. select the standard of physical server when moving: all be that the utilance of physical server by that attribute that surpasses maximum disparity sorted during migration at every turn, all be the physical server of selecting the utilance maximum then, and then select virtual machine from this physical server.
8. select the standard of virtual machine when moving: after finding the physical server that needs migration, find the virtual machine of utilance minimum to move again from above, can guarantee the success rate of moving so to greatest extent, be placed on the another one physical server because if move the virtual machine of utilance maximum at every turn, another one physical server operating availability also can be excessive.The little virtual machine of each migration utilance can be avoided this situation.
9. move to the standard of any platform physical server: then directly call the physical server that existing allocation algorithm searching should be moved to after finding the virtual machine that needs migration.
10. the condition that stops of migration: when peak use rate and minimum utilance difference stop migration during less than the requiring of maximum disparity, or occurred repeatedly till the situation of migration.
11. the load balancing device, the load balancing device shown in Figure of description 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100903353A CN102185759A (en) | 2011-04-12 | 2011-04-12 | Multi-physical server load equalizing method and device capable of meeting requirement characteristic |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011100903353A CN102185759A (en) | 2011-04-12 | 2011-04-12 | Multi-physical server load equalizing method and device capable of meeting requirement characteristic |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102185759A true CN102185759A (en) | 2011-09-14 |
Family
ID=44571830
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011100903353A Pending CN102185759A (en) | 2011-04-12 | 2011-04-12 | Multi-physical server load equalizing method and device capable of meeting requirement characteristic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102185759A (en) |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102404412A (en) * | 2011-12-28 | 2012-04-04 | 北京邮电大学 | Energy saving method and system for cloud compute data center |
| CN102521055A (en) * | 2011-12-15 | 2012-06-27 | 广州杰赛科技股份有限公司 | Virtual machine resource allocating method and virtual machine resource allocating system |
| CN102629216A (en) * | 2012-02-24 | 2012-08-08 | 浪潮(北京)电子信息产业有限公司 | Cloud operating system (OS) scheduling method and cloud system scheduling device |
| CN102917014A (en) * | 2012-09-10 | 2013-02-06 | 中国联合网络通信集团有限公司 | Resource scheduling method and device |
| CN102929701A (en) * | 2012-10-17 | 2013-02-13 | 北京搜狐新媒体信息技术有限公司 | Scheduling method and device for batched virtual machines |
| CN102932279A (en) * | 2012-10-30 | 2013-02-13 | 北京邮电大学 | Multidimensional resource scheduling system and method for cloud environment data center |
| CN102984082A (en) * | 2012-11-30 | 2013-03-20 | 刘庆国 | Network service quality control method and device |
| CN103036979A (en) * | 2012-12-12 | 2013-04-10 | 广州尚融网络科技有限公司 | Server loading balancing method and loading balancer |
| WO2013123650A1 (en) * | 2012-02-22 | 2013-08-29 | 华为技术有限公司 | Method for virtual machine assignment and device for virtual machine assignment |
| CN103327093A (en) * | 2013-06-17 | 2013-09-25 | 苏州市职业大学 | Method for controlling cloud computing system |
| CN103514046A (en) * | 2013-09-24 | 2014-01-15 | 华为技术有限公司 | Virtual machine placement method and cluster management server |
| CN103576827A (en) * | 2012-07-25 | 2014-02-12 | 田文洪 | Method and device of online energy-saving dispatching in cloud computing data center |
| CN103763346A (en) * | 2013-12-31 | 2014-04-30 | 华为技术有限公司 | Distributed resource scheduling method and device |
| CN103905472A (en) * | 2012-12-25 | 2014-07-02 | 中国移动通信集团江苏有限公司 | Cloud storage method and system for access through webpage mode |
| CN103973586A (en) * | 2014-05-14 | 2014-08-06 | 江苏省无线电科学研究所有限公司 | Method for achieving automatic meteorological station wireless networking load balance |
| CN104375897A (en) * | 2014-10-27 | 2015-02-25 | 西安工程大学 | Cloud computing resource scheduling method based on minimum relative load imbalance degree |
| CN105007337A (en) * | 2015-08-20 | 2015-10-28 | 浪潮(北京)电子信息产业有限公司 | Cluster system load balancing method and system thereof |
| CN105141656A (en) * | 2015-07-20 | 2015-12-09 | 浙江工商大学 | Internet lightweight application load balancing realization method based on cloud platforms |
| CN105159784A (en) * | 2015-10-10 | 2015-12-16 | 浪潮(北京)电子信息产业有限公司 | Deployment method and system for virtual servers |
| CN105183670A (en) * | 2015-10-27 | 2015-12-23 | 北京百度网讯科技有限公司 | Data processing method and device used for distributed cache system |
| CN105472018A (en) * | 2015-12-22 | 2016-04-06 | 曙光信息产业股份有限公司 | Flow detection method, load balancer, detection server and flow detection system |
| CN105893155A (en) * | 2016-04-14 | 2016-08-24 | 广州杰赛科技股份有限公司 | Virtual machine load balancing control method and device |
| CN106453564A (en) * | 2016-10-18 | 2017-02-22 | 北京京东尚科信息技术有限公司 | Elastic cloud distributed massive request processing method, device and system |
| CN107273185A (en) * | 2017-06-19 | 2017-10-20 | 成都鼎智汇科技有限公司 | A kind of control method for equalizing load based on virtual machine |
| CN108170517A (en) * | 2018-01-08 | 2018-06-15 | 武汉斗鱼网络科技有限公司 | A kind of container allocation method, apparatus, server and medium |
| CN108519919A (en) * | 2018-03-19 | 2018-09-11 | 山东超越数控电子股份有限公司 | A method of realizing server resource dynamic dispatching under virtual cluster environment |
| CN109495542A (en) * | 2018-10-16 | 2019-03-19 | 深圳壹账通智能科技有限公司 | Load allocation method and terminal device based on performance monitoring |
| CN110198359A (en) * | 2019-07-08 | 2019-09-03 | 紫光云技术有限公司 | A kind of load-balancing method and device |
| CN111953732A (en) * | 2015-11-10 | 2020-11-17 | 华为技术有限公司 | Resource scheduling method and device in cloud computing system |
| CN113946436A (en) * | 2021-07-29 | 2022-01-18 | 西北大学 | Resource pre-scheduling method based on load balancing |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101309208A (en) * | 2008-06-21 | 2008-11-19 | 华中科技大学 | Job scheduling system suitable for grid environment and based on reliable expense |
| CN101639793A (en) * | 2009-08-19 | 2010-02-03 | 南京邮电大学 | Grid load predicting method based on support vector regression machine |
| US7669029B1 (en) * | 2006-11-15 | 2010-02-23 | Network Appliance, Inc. | Load balancing a data storage system |
-
2011
- 2011-04-12 CN CN2011100903353A patent/CN102185759A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7669029B1 (en) * | 2006-11-15 | 2010-02-23 | Network Appliance, Inc. | Load balancing a data storage system |
| CN101309208A (en) * | 2008-06-21 | 2008-11-19 | 华中科技大学 | Job scheduling system suitable for grid environment and based on reliable expense |
| CN101639793A (en) * | 2009-08-19 | 2010-02-03 | 南京邮电大学 | Grid load predicting method based on support vector regression machine |
Non-Patent Citations (1)
| Title |
|---|
| 郑洪源等: "WEB服务器集群系统中负载平衡的设计与实现", 《南京航空航天大学学报》, vol. 38, no. 3, 30 June 2006 (2006-06-30), pages 347 - 351 * |
Cited By (50)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521055A (en) * | 2011-12-15 | 2012-06-27 | 广州杰赛科技股份有限公司 | Virtual machine resource allocating method and virtual machine resource allocating system |
| CN102404412B (en) * | 2011-12-28 | 2014-01-08 | 北京邮电大学 | Energy saving method and system for cloud compute data center |
| CN102404412A (en) * | 2011-12-28 | 2012-04-04 | 北京邮电大学 | Energy saving method and system for cloud compute data center |
| CN103548324A (en) * | 2012-02-22 | 2014-01-29 | 华为技术有限公司 | Method for virtual machine assignment and device for virtual machine assignment |
| WO2013123650A1 (en) * | 2012-02-22 | 2013-08-29 | 华为技术有限公司 | Method for virtual machine assignment and device for virtual machine assignment |
| CN103548324B (en) * | 2012-02-22 | 2016-06-15 | 华为技术有限公司 | Virtual machine distribution method and virtual machine distributor |
| CN102629216A (en) * | 2012-02-24 | 2012-08-08 | 浪潮(北京)电子信息产业有限公司 | Cloud operating system (OS) scheduling method and cloud system scheduling device |
| CN103576827A (en) * | 2012-07-25 | 2014-02-12 | 田文洪 | Method and device of online energy-saving dispatching in cloud computing data center |
| CN103576827B (en) * | 2012-07-25 | 2016-04-27 | 田文洪 | The method of the online energy-saving distribution of a kind of cloud computation data center |
| CN102917014A (en) * | 2012-09-10 | 2013-02-06 | 中国联合网络通信集团有限公司 | Resource scheduling method and device |
| CN102917014B (en) * | 2012-09-10 | 2015-11-25 | 中国联合网络通信集团有限公司 | Resource regulating method and device |
| CN102929701B (en) * | 2012-10-17 | 2015-04-29 | 北京搜狐新媒体信息技术有限公司 | Scheduling method and device for batched virtual machines |
| CN102929701A (en) * | 2012-10-17 | 2013-02-13 | 北京搜狐新媒体信息技术有限公司 | Scheduling method and device for batched virtual machines |
| CN102932279B (en) * | 2012-10-30 | 2016-06-29 | 北京邮电大学 | A kind of cloud environment data center multi dimensional resource dispatching method |
| CN102932279A (en) * | 2012-10-30 | 2013-02-13 | 北京邮电大学 | Multidimensional resource scheduling system and method for cloud environment data center |
| CN102984082A (en) * | 2012-11-30 | 2013-03-20 | 刘庆国 | Network service quality control method and device |
| CN102984082B (en) * | 2012-11-30 | 2015-06-03 | 温州大学 | Network service quality control method and device |
| CN103036979A (en) * | 2012-12-12 | 2013-04-10 | 广州尚融网络科技有限公司 | Server loading balancing method and loading balancer |
| CN103036979B (en) * | 2012-12-12 | 2015-07-01 | 广州尚融网络科技有限公司 | Server loading balancing method and loading balancer |
| CN103905472A (en) * | 2012-12-25 | 2014-07-02 | 中国移动通信集团江苏有限公司 | Cloud storage method and system for access through webpage mode |
| CN103905472B (en) * | 2012-12-25 | 2017-12-01 | 中国移动通信集团江苏有限公司 | A kind of cloud storage method and system accessed by webpage mode |
| CN103327093A (en) * | 2013-06-17 | 2013-09-25 | 苏州市职业大学 | Method for controlling cloud computing system |
| CN103327093B (en) * | 2013-06-17 | 2016-04-27 | 苏州市职业大学 | The control method of cloud computing system |
| CN103514046A (en) * | 2013-09-24 | 2014-01-15 | 华为技术有限公司 | Virtual machine placement method and cluster management server |
| CN103763346A (en) * | 2013-12-31 | 2014-04-30 | 华为技术有限公司 | Distributed resource scheduling method and device |
| CN103973586A (en) * | 2014-05-14 | 2014-08-06 | 江苏省无线电科学研究所有限公司 | Method for achieving automatic meteorological station wireless networking load balance |
| CN104375897A (en) * | 2014-10-27 | 2015-02-25 | 西安工程大学 | Cloud computing resource scheduling method based on minimum relative load imbalance degree |
| CN104375897B (en) * | 2014-10-27 | 2018-02-27 | 西安工程大学 | Cloud computing resource scheduling method based on the unbalanced degree of minimum relative load |
| CN105141656B (en) * | 2015-07-20 | 2018-05-01 | 浙江工商大学 | A kind of implementation of load balancing of the internet lightweight application based on cloud platform |
| CN105141656A (en) * | 2015-07-20 | 2015-12-09 | 浙江工商大学 | Internet lightweight application load balancing realization method based on cloud platforms |
| CN105007337A (en) * | 2015-08-20 | 2015-10-28 | 浪潮(北京)电子信息产业有限公司 | Cluster system load balancing method and system thereof |
| CN105159784A (en) * | 2015-10-10 | 2015-12-16 | 浪潮(北京)电子信息产业有限公司 | Deployment method and system for virtual servers |
| CN105159784B (en) * | 2015-10-10 | 2018-07-06 | 浪潮(北京)电子信息产业有限公司 | The dispositions method and system of a kind of virtual server |
| CN105183670A (en) * | 2015-10-27 | 2015-12-23 | 北京百度网讯科技有限公司 | Data processing method and device used for distributed cache system |
| CN105183670B (en) * | 2015-10-27 | 2018-11-27 | 北京百度网讯科技有限公司 | Data processing method and device for distributed cache system |
| CN111953732B (en) * | 2015-11-10 | 2023-03-28 | 华为技术有限公司 | Resource scheduling method and device in cloud computing system |
| CN111953732A (en) * | 2015-11-10 | 2020-11-17 | 华为技术有限公司 | Resource scheduling method and device in cloud computing system |
| CN105472018A (en) * | 2015-12-22 | 2016-04-06 | 曙光信息产业股份有限公司 | Flow detection method, load balancer, detection server and flow detection system |
| CN105893155A (en) * | 2016-04-14 | 2016-08-24 | 广州杰赛科技股份有限公司 | Virtual machine load balancing control method and device |
| CN106453564B (en) * | 2016-10-18 | 2019-11-05 | 北京京东尚科信息技术有限公司 | The method, apparatus and system of elastic cloud distribution magnanimity request processing |
| CN106453564A (en) * | 2016-10-18 | 2017-02-22 | 北京京东尚科信息技术有限公司 | Elastic cloud distributed massive request processing method, device and system |
| CN107273185B (en) * | 2017-06-19 | 2020-09-29 | 观典防务技术股份有限公司 | Load balancing control method based on virtual machine |
| CN107273185A (en) * | 2017-06-19 | 2017-10-20 | 成都鼎智汇科技有限公司 | A kind of control method for equalizing load based on virtual machine |
| CN108170517A (en) * | 2018-01-08 | 2018-06-15 | 武汉斗鱼网络科技有限公司 | A kind of container allocation method, apparatus, server and medium |
| CN108519919A (en) * | 2018-03-19 | 2018-09-11 | 山东超越数控电子股份有限公司 | A method of realizing server resource dynamic dispatching under virtual cluster environment |
| CN109495542A (en) * | 2018-10-16 | 2019-03-19 | 深圳壹账通智能科技有限公司 | Load allocation method and terminal device based on performance monitoring |
| CN109495542B (en) * | 2018-10-16 | 2022-08-02 | 深圳壹账通智能科技有限公司 | Load distribution method based on performance monitoring and terminal equipment |
| CN110198359A (en) * | 2019-07-08 | 2019-09-03 | 紫光云技术有限公司 | A kind of load-balancing method and device |
| CN113946436A (en) * | 2021-07-29 | 2022-01-18 | 西北大学 | Resource pre-scheduling method based on load balancing |
| CN113946436B (en) * | 2021-07-29 | 2024-03-08 | 西北大学 | Resource pre-scheduling method based on load balancing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102185759A (en) | Multi-physical server load equalizing method and device capable of meeting requirement characteristic | |
| US7945913B2 (en) | Method, system and computer program product for optimizing allocation of resources on partitions of a data processing system | |
| CN102232282B (en) | Method and apparatus for realizing load balance of resources in data center | |
| EP3522500B1 (en) | Method and system for multi-tenant resource distribution | |
| CN102185779B (en) | Method and device for realizing data center resource load balance in proportion to comprehensive allocation capability | |
| CN108182105B (en) | Local dynamic migration method and control system based on Docker container technology | |
| CN102130938A (en) | Resource supply method oriented to Web application host platform | |
| US9846600B1 (en) | System and method to dynamically allocate varying processing capacity entitlements based on workload importance | |
| US9823875B2 (en) | Transparent hybrid data storage | |
| US8984521B2 (en) | Computer system performance by applying rate limits to control block tenancy | |
| JP2012522320A (en) | System and method for managing priority-based system load levels | |
| CN103793272A (en) | Periodical task scheduling method and periodical task scheduling system | |
| WO2015101419A1 (en) | Method and system for allocating resources to resource consumers in a cloud computing environment | |
| US20170339069A1 (en) | Allocating Cloud Computing Resources In A Cloud Computing Environment | |
| CN102339233A (en) | Cloud computing centralized management platform | |
| CN105430027A (en) | Load balance dynamic pre-allocating method based on a plurality of resource scales | |
| US10643193B2 (en) | Dynamic workload capping | |
| CN112463395A (en) | Resource allocation method, device, equipment and readable storage medium | |
| US10812278B2 (en) | Dynamic workload capping | |
| AU2021103249A4 (en) | A novel multi-level optimization for task scheduling and load balancing in cloud | |
| Bhatia et al. | Perspective study on load balancing paradigms in cloud computing | |
| Shu-Jun et al. | Optimization and research of hadoop platform based on fifo scheduler | |
| Gadhavi et al. | Efficient resource provisioning through workload prediction in the cloud system | |
| Narang et al. | Various load balancing techniques in cloud computing | |
| Perron et al. | Cackle: Analytical Workload Cost and Performance Stability With Elastic Pools |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C53 | Correction of patent of invention or patent application | ||
| CB02 | Change of applicant information |
Address after: Chenghua District Chengdu city Sichuan province 610054 Taoxi Road No. 189 building 4 unit 1 Taoxi Yiyuan No. 10 Applicant after: Tian Wenhong Address before: Sichuan city of Chengdu province Chenghua District 610054 foot Water Bridge No. 2 North Street 38 No. 506 dianzaikeda Dongyuan real Applicant before: Tian Wenhong |
|
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110914 |
|
| RJ01 | Rejection of invention patent application after publication |