CN109800059A - A kind of cloud computing virtual machine migration method based on load curve similarity - Google Patents
A kind of cloud computing virtual machine migration method based on load curve similarity Download PDFInfo
- Publication number
- CN109800059A CN109800059A CN201910094908.6A CN201910094908A CN109800059A CN 109800059 A CN109800059 A CN 109800059A CN 201910094908 A CN201910094908 A CN 201910094908A CN 109800059 A CN109800059 A CN 109800059A
- Authority
- CN
- China
- Prior art keywords
- resource
- virtual machine
- physical machine
- machine
- matrix
- 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
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention proposes a kind of cloud computing virtual machine migration methods based on load curve similarity, obtain the surplus resources utilization rate curve of all physical machines and the resource utilization curve of all virtual machines to be migrated, according to Pearson correlation coefficients formula, obtain the similarity vector being made of the similarity of every a kind of resource, Euclidean distance is finally asked according to similarity vector and average resource, to determine virtual machine and target physical machine that needs migrate.The present invention will overload standard of the physical machine load interior for a period of time as migration, rather than it is concerned only with the flashy loading condition of overload, it is effectively guaranteed after migration that target physical machine will not overload in a period of time, to reduce migration number, reduces energy consumption.Meanwhile using the similarity of curve as one of standard of migration, keep the virtual machine that must be migrated and target physical machine complementary as far as possible, so that the load of target physical machine is more balanced.
Description
Technical field
The invention belongs to cloud computings and virtualization related fields, and in particular to a kind of based on the cloud of load curve similarity
Calculate virtual machine migration method.
Background technique
The initial target that cloud computing is suggested is mainly to manage computing resource, network to reinforce the management to resource
Three aspects of resource and storage resource.The main method strengthened management is to improve the flexibility of time and space.Time it is flexible
Property, i.e., can offer service at any time when user needs.The flexibility in space, i.e., according to resource needed for every user
Amount is to provide service.The flexibility of time and space combine, as the elasticity of cloud computing.
In order to improve the elasticity of cloud computing, it has been suggested that excessively many solutions.Initial solution is to provide physics
The ability of equipment, i.e. raising hard disc data amount of storage, server memory, increase device network bandwidth etc..But this method has
The limitation of itself, it can not thoroughly solve this requirement of flexibility.Since the needs such as server, the network equipment are spent largely
Time is purchased, it is impossible to be obtained at any time.
Then hardware " virtualization " technology is come into being.One key characteristic of virtualization technology is to migrate online: one
The virtual machine being currently running moves to an other physical host from the physical host at place.
It is frequently necessary to migrate the physical host of overload in real life, but many real-time migration algorithms are only examined
The resource service condition for having considered this virtual machine in physical machine in a flash of overload, carrys out selection target physical machine.But it is different virtual
The resource utilization of machine would generally not stop to change at any time, this is possible to cause the virtual machine (vm) migration to one section of destination host
The overload for causing destination host after time again needs to migrate again, has both increased the number of migration, also increases energy consumption, drop
Low virtual machine performance.
Summary of the invention
The present invention is in view of the above-mentioned problems, propose a kind of cloud computing virtual machine (vm) migration side based on load curve similarity
Method, main thought are: obtaining the surplus resources utilization rate curve of all physical machines and the resource benefit of all virtual machines to be migrated
The similarity vector being made of the similarity of every a kind of resource is obtained according to Pearson correlation coefficients formula with rate curve.Finally
Euclidean distance is sought according to similarity vector and average resource, to determine virtual machine and target physical machine that needs migrate.
A kind of cloud computing virtual machine migration method based on load curve similarity, characterized by the following steps:
Step 1, it defines physical machine set and overloads the virtual machine set in physical machine, while defining what physical machine was possessed
Resource；
Step 2, the utilization of resources rate matrix for obtaining each virtual machine in overload physical machine, establishes resource utilization matrix stack
It closes, indicates the set of the utilization of resources rate matrix of all virtual machines being located in overload physical machine；
Step 3, each virtual machine is calculated in a period of time to the mean value and standard deviation of the utilization rate of resource, so
All resources of virtual machine in resource utilization set of matrices are standardized using rate matrix afterwards, obtain normalized matrix；
Step 4, the utilization of resources rate matrix in each physical machine same amount of time is obtained；
Step 5, the residue for obtaining resource in physical machine utilizes rate matrix；
Step 6, each physical machine is calculated in a period of time to the surplus resources utilization rate mean value and standard of resource
The surplus resources of physical machine are standardized using rate matrix, obtain normalized matrix by difference；
Step 7, Pearson's phase is asked respectively using every row of rate matrix with surplus resources to every row of utilization of resources rate matrix
Relationship number obtains the similarity vector about every a kind of resource；
Step 8, according to similarity vector, the maximum of the similarity of resource between all virtual machines and physical machine pair is obtained
Value；
Step 9, the Euclidean distance for calculating every a pair of of virtual machine and physical machine, obtains Euclidean distance set；
Step 10, in Euclidean distance set, corresponding overload virtual machine and physics when Euclidean distance is minimized are obtained
Machine, it is assumed that will overload on the virtual machine (vm) migration to the physical machine, whether meet after judging migration, physical machine is in certain time pair
Whether the utilization rate of resource is less than the threshold value of resource utilization in physical machine, if it is satisfied, then by virtual machine (vm) migration to physical machine
On, and corresponding matrix is removed from resource utilization set of matrices；Otherwise by the Euclidean distance from Euclidean distance set
It removes, re-starts step 10；
Step 11, if overload virtual machine no longer overloads or resources of virtual machine utilization rate set of matrices is sky, terminate to move
It moves；If it is not, then return step 2, reselects the virtual machine to be migrated and target physical machine.
Further, the step 1 defines physical machine set P={ p1,p2,…,pj,…,pn, n platform is shared in set
Physical machine, wherein j indicates jth platform physical machine, overloads the virtual machine set V={ v in physical machine1,v2,…vi…,vm, set V
In share m platform virtual machine, wherein i indicates i-th virtual machine in set V, and physical machine possesses k class resource altogether, including CPU, memory,
Hard disk etc..
Further, the step 2 obtains the resource utilization matrix A of each virtual machine in overload physical machinei；
Wherein uitkIt is defined as virtual machine viIn time t to the utilization rate of resource k, wherein every a line represent it is same virtual
Machine viFrom the time 1 to the time to the utilization rate of resource k when t, each column represent a kind of different resource；
Resource utilization set of matrices S={ A1,…,Ai,…,Am, indicate all virtual machines being located in overload physical machine
Utilization of resources rate matrix set.
Further, the step 3 calculates each virtual machine v according to formula (1)iTo the benefit of resource k in time t
With rate mean value
Each virtual machine v is calculated according to formula (2)iTo the standard deviation sigma of the utilization rate of resource k in time tik；
All resources of virtual machine utilization rate matrix A i in resource utilization set of matrices S are standardized according to formula (3),
Obtain normalized matrix Xi；
Thus it is obtained and carries each virtual machine v in physical machineiResource utilization normalized matrix Xi,。
Further, the step 4 obtains each physical machine pjResource utilization matrix B in same amount of time tj；
Wherein UjtkIt is defined as physical machine pjIn time t, to the utilization rate of resource k, wherein every a line represents same physics
Machine pjFrom the time 1 to the time to the utilization rate of resource k when t, each column represent a kind of different resource.
Further, the step 5 obtains physical machine p according to formula (4)jThe remaining utilization rate matrix B of upper resource k 'j,
Wherein TjkIndicate physical machine pjThe threshold value of upper resource k utilization rate:
Further, the step 6 calculates each physical machine p according to formula (5)jResource k is remained in time t
Remaining resource utilization mean value
Each physical machine p is calculated according to formula (6)jTo the standard deviation of the surplus resources utilization rate of resource k in time t
σ'jk；
According to formula (7) by physical machine pjSurplus resources utilization rate matrix B 'jIt is standardized, obtains normalized matrix
X'j:
Thus each physical machine p has been respectively obtainedjSurplus resources utilization rate normalized matrix X 'j,。
Further, the step 7, according to formula (8), by matrix XiEvery row and matrix X'jEvery row acquire Pearson
Related coefficient hijk；
Obtain virtual machine viWith physical machine pjSimilarity vector H about all resourcesij={ hij1,hij2,…,hijk,
Middle hijkIndicate virtual machine viResource k utilization rate curve and physical machine pjResource k remaining utilization rate curve similarity.
Further, the step 8 obtains ideal similarity vector H according to formula (9)*, H*In each rkIt represents
The maximum value of the similarity of resource k between all virtual machines and physical machine pair；
H*={ max (p111,p121,…,pij1),…,max(p11k,p12k,…,pijk)={ r1,r2,…,rk} (9)
Further, the step 9 calculates every a pair of of virtual machine v according to formula (10)iWith physical machine pjEuclidean distance
lij；
Euclidean distance set L indicates the Euclidean distance l of all virtual machine physical machines pairijSet, L={ l11,…,
l1j,…,lij}。
A kind of virtual machine migration policies the invention proposes cloud computing based on load curve similarity, more current mainstream
For virtual machine migration policies, the main advantage of the strategy is: will overload physical machine load interior for a period of time as migration
Standard, rather than be concerned only with the flashy loading condition of overload, object in a period of time be effectively guaranteed after migration
Reason machine will not overload, to reduce migration number, reduce energy consumption.Meanwhile using the similarity of curve as migration
One of standard keeps the virtual machine that must be migrated and target physical machine complementary as far as possible, so that the load of target physical machine
It is more balanced.
Detailed description of the invention
Fig. 1 is a kind of cloud computing virtual machine migration method flow chart based on load curve similarity of the present invention.
Specific embodiment
Technical solution of the present invention is described in further detail with reference to the accompanying drawings of the specification.
A kind of cloud computing virtual machine migration method based on load curve similarity, includes the following steps:
Step 1, physical machine set P={ p is defined1,p2,…,pj,…,pn, n platform physical machine is shared in set, wherein j table
Show jth platform physical machine, overloads the virtual machine set V={ v in physical machine1,v2,…vi…,vm, it is virtual that m platform is shared in set V
Machine, wherein i indicates i-th virtual machine in set V, and physical machine possesses k class resource, including CPU, memory, hard disk etc. altogether.
Step 2, the resource utilization matrix A of each virtual machine in overload physical machine is obtainedi。
Wherein uitkIt is defined as virtual machine viIn time t to the utilization rate of resource k, wherein every a line represent it is same virtual
Machine viFrom the time 1 to the time to the utilization rate of resource k when t, each column represent a kind of different resource.
Resource utilization set of matrices S={ A1,…,Ai,…,Am, indicate all virtual machines being located in overload physical machine
Utilization of resources rate matrix set.
Step 3, each virtual machine v is calculated according to formula (1)iTo the utilization rate mean value of resource k in time t
Each virtual machine v is calculated according to formula (2)iTo the standard deviation sigma of the utilization rate of resource k in time tik。
All resources of virtual machine utilization rate matrix A i in resource utilization set of matrices S are standardized according to formula (3),
Obtain normalized matrix Xi。
Thus it is obtained and carries each virtual machine v in physical machineiResource utilization normalized matrix Xi。
Step 4, each physical machine p is obtainedjResource utilization matrix B in same amount of time tj。
Wherein UjtkIt is defined as physical machine pjIn time t, to the utilization rate of resource k, wherein every a line represents same physics
Machine pjFrom the time 1 to the time to the utilization rate of resource k when t, each column represent a kind of different resource.
Step 5, physical machine p is obtained according to formula (4)jThe remaining utilization rate matrix B of upper resource k 'j, wherein TjkExpression thing
Reason machine pjThe threshold value of upper resource k utilization rate:
Step 6, each physical machine p is calculated according to formula (5)jIt is equal to the surplus resources utilization rate of resource k in time t
Value
Each physical machine p is calculated according to formula (6)jTo the standard deviation of the surplus resources utilization rate of resource k in time t
σ'jk。
According to formula (7) by physical machine pjSurplus resources utilization rate matrix B 'jIt is standardized, obtains normalized matrix
X'j。
Thus each physical machine p has been respectively obtainedjSurplus resources utilization rate normalized matrix X 'j。
Step 7, according to formula (8), by matrix XiEvery row and matrix X'jEvery row acquire Pearson correlation coefficients hijk。
Obtain virtual machine viWith physical machine pjSimilarity vector H about all resourcesij={ hij1,hij2,…,hijk,
Middle hijkIndicate virtual machine viResource k utilization rate curve and physical machine pjResource k remaining utilization rate curve similarity.
Step 8, ideal similarity vector H is obtained according to formula (9)*, H*In each rkRepresent all virtual machines and object
The maximum value of the similarity of resource k between reason machine pair.
H*={ max (p111,p121,…,pij1),…,max(p11k,p12k,…,pijk)={ r1,r2,…,rk} (9)
Step 9, every a pair of of virtual machine v is calculated according to formula (10)iWith physical machine pjEuclidean distance lij。
Euclidean distance set L indicates the Euclidean distance l of all virtual machine physical machines pairijSet, L={ l11,…,
l1j,…,lij}。
Step 10, in Euclidean distance set L, l is worked as in acquisitionijSubscript i and j when being minimized；Assuming that by virtual machine vi
Move to physical machine pjOn, whether meet after judging migrationIf it is satisfied, then by virtual machine viMove to object
Reason machine pjOn, and A is removed from resource utilization set of matrices Si；Otherwise by lijIt is removed from Euclidean distance set L, again
Carry out step 10.
Step 11, if overload virtual machine PjNo longer overload or resources of virtual machine utilization rate set of matrices S is sky, then terminates
Migration；If it is not, then return step 2, reselects the virtual machine to be migrated and target physical machine.
The foregoing is merely better embodiment of the invention, protection scope of the present invention is not with above embodiment
Limit, as long as those of ordinary skill in the art's equivalent modification or variation made by disclosure according to the present invention, should all be included in power
In the protection scope recorded in sharp claim.
Claims (10)
1. a kind of cloud computing virtual machine migration method based on load curve similarity, characterized by the following steps:
Step 1, it defines physical machine set and overloads the virtual machine set in physical machine, while defining the money that physical machine is possessed
Source；
Step 2, the utilization of resources rate matrix for obtaining each virtual machine in overload physical machine, establishes resource utilization set of matrices,
Indicate the set of the utilization of resources rate matrix of all virtual machines being located in overload physical machine；
Step 3, each virtual machine is calculated in a period of time to the mean value and standard deviation of the utilization rate of resource, it is then right
All resources of virtual machine in resource utilization set of matrices are standardized using rate matrix, obtain normalized matrix；
Step 4, the utilization of resources rate matrix in each physical machine same amount of time is obtained；
Step 5, the residue for obtaining resource in physical machine utilizes rate matrix；
Step 6, each physical machine is calculated in a period of time to the surplus resources utilization rate mean value and standard deviation of resource, is incited somebody to action
The surplus resources of physical machine are standardized using rate matrix, obtain normalized matrix；
Step 7, pearson correlation system is asked respectively using every row of rate matrix to every row of utilization of resources rate matrix and surplus resources
Number obtains the similarity vector about every a kind of resource；
Step 8, according to similarity vector, the maximum value of the similarity of resource between all virtual machines and physical machine pair is obtained；
Step 9, the Euclidean distance for calculating every a pair of of virtual machine and physical machine, obtains Euclidean distance set；
Step 10, in Euclidean distance set, corresponding overload virtual machine and physical machine when Euclidean distance is minimized are obtained, it is false
If will overload on the virtual machine (vm) migration to the physical machine, whether meet after judging migration, physical machine is in certain time to resource
Utilization rate whether be less than the threshold value of resource utilization in physical machine, if it is satisfied, then by virtual machine (vm) migration to physical machine,
And corresponding matrix is removed from resource utilization set of matrices；Otherwise the Euclidean distance is removed from Euclidean distance set, weight
It is new to carry out step 10；
Step 11, if overload virtual machine no longer overloads or resources of virtual machine utilization rate set of matrices is sky, terminate to migrate；Such as
Fruit is not that then return step 2, reselect the virtual machine to be migrated and target physical machine.
2. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
Be: the step 1 defines physical machine set P={ p1,p2,…,pj,…,pn, n platform physical machine is shared in set, wherein j
It indicates jth platform physical machine, overloads the virtual machine set V={ v in physical machine1,v2,…vi…,vm, it is empty that m platform is shared in set V
Quasi- machine, wherein i indicates i-th virtual machine in set V, and physical machine possesses k class resource, including CPU, memory, hard disk etc. altogether.
3. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
Be: the step 2 obtains the resource utilization matrix A of each virtual machine in overload physical machinei；
Wherein uitkIt is defined as virtual machine viIn time t to the utilization rate of resource k, wherein every a line represents same virtual machine viFrom
To the utilization rate of resource k when time 1 to time t, each column represent a kind of different resource；
Resource utilization set of matrices S={ A1,…,Ai,…,Am, indicate the money of all virtual machines being located in overload physical machine
The set of source utilization rate matrix.
4. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
Be: the step 3 calculates each virtual machine v according to formula (1)iTo the utilization rate mean value of resource k in time t
Each virtual machine v is calculated according to formula (2)iTo the standard deviation sigma of the utilization rate of resource k in time tik；
According to formula (3) to all resources of virtual machine utilization rate matrix As in resource utilization set of matrices SiStandardization, obtains
Normalized matrix Xi；
Thus it is obtained and carries each virtual machine v in physical machineiResource utilization normalized matrix Xi,。
5. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
Be: the step 4 obtains each physical machine pjResource utilization matrix B in same amount of time tj；
Wherein UjtkIt is defined as physical machine pjIn time t, to the utilization rate of resource k, wherein every a line represents same physical machine pj
From the time 1 to the time to the utilization rate of resource k when t, each column represent a kind of different resource.
6. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
Be: the step 5 obtains physical machine p according to formula (4)jThe remaining utilization rate matrix B of upper resource k 'j, wherein TjkExpression thing
Reason machine pjThe threshold value of upper resource k utilization rate:
7. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
Be: the step 6 calculates each physical machine p according to formula (5)jTo the surplus resources utilization rate of resource k in time t
Mean value
Each physical machine p is calculated according to formula (6)jIn time t to the standard deviation sigma of the surplus resources utilization rate of resource k 'jk；
According to formula (7) by physical machine pjSurplus resources utilization rate matrix B 'jIt is standardized, obtains normalized matrix X'j:
Thus each physical machine p has been respectively obtainedjSurplus resources utilization rate normalized matrix X 'j,。
8. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
It is: the step 7, according to formula (8), by matrix XiEvery row and matrix X'jEvery row acquire Pearson correlation coefficients hijk；
Obtain virtual machine viWith physical machine pjSimilarity vector H about all resourcesij={ hij1,hij2,…,hijk, wherein hijk
Indicate virtual machine viResource k utilization rate curve and physical machine pjResource k remaining utilization rate curve similarity.
9. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
Be: the step 8 obtains ideal similarity vector H according to formula (9)*, H*In each rkRepresent all virtual machines with
The maximum value of the similarity of resource k between physical machine pair；
H*={ max (p111,p121,…,pij1),…,max(p11k,p12k,…,pijk)={ r1,r2,…,rk}(9)
10. a kind of cloud computing virtual machine migration method based on load curve similarity according to claim 1, feature
Be: the step 9 calculates every a pair of of virtual machine v according to formula (10)iWith physical machine pjEuclidean distance lij；
Euclidean distance set L indicates the Euclidean distance l of all virtual machine physical machines pairijSet, L={ l11,…,l1j,…,
lij}。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910094908.6A CN109800059B (en) | 2019-01-31 | 2019-01-31 | Load curve similarity-based cloud computing virtual machine migration method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910094908.6A CN109800059B (en) | 2019-01-31 | 2019-01-31 | Load curve similarity-based cloud computing virtual machine migration method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109800059A true CN109800059A (en) | 2019-05-24 |
| CN109800059B CN109800059B (en) | 2022-10-21 |
Family
ID=66559194
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910094908.6A Active CN109800059B (en) | 2019-01-31 | 2019-01-31 | Load curve similarity-based cloud computing virtual machine migration method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109800059B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110399203A (en) * | 2019-07-25 | 2019-11-01 | 南京邮电大学 | A kind of cloud computing virtual machine migration method based on Fu Leixie distance |
| WO2021180056A1 (en) * | 2020-03-09 | 2021-09-16 | 华为技术有限公司 | Method for resource migration, system and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130145364A1 (en) * | 2011-12-01 | 2013-06-06 | Tunghai University | Method of dynamic resource allocation for a virtual machine cluster |
| CN105279027A (en) * | 2015-11-19 | 2016-01-27 | 浪潮(北京)电子信息产业有限公司 | Virtual machine disposition method and apparatus |
| CN106293871A (en) * | 2016-07-22 | 2017-01-04 | 浪潮(北京)电子信息产业有限公司 | A kind of resource regulating method of cluster virtual machine |
-
2019
- 2019-01-31 CN CN201910094908.6A patent/CN109800059B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130145364A1 (en) * | 2011-12-01 | 2013-06-06 | Tunghai University | Method of dynamic resource allocation for a virtual machine cluster |
| CN105279027A (en) * | 2015-11-19 | 2016-01-27 | 浪潮(北京)电子信息产业有限公司 | Virtual machine disposition method and apparatus |
| CN106293871A (en) * | 2016-07-22 | 2017-01-04 | 浪潮(北京)电子信息产业有限公司 | A kind of resource regulating method of cluster virtual machine |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110399203A (en) * | 2019-07-25 | 2019-11-01 | 南京邮电大学 | A kind of cloud computing virtual machine migration method based on Fu Leixie distance |
| WO2021180056A1 (en) * | 2020-03-09 | 2021-09-16 | 华为技术有限公司 | Method for resource migration, system and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109800059B (en) | 2022-10-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10841241B2 (en) | Intelligent placement within a data center | |
| CN104252393B (en) | Manage the methods and applications resource distribution module of resource in Distributed Computer System | |
| CN107196865A (en) | A kind of adaptive threshold overload moving method of Load-aware | |
| CN109710374A (en) | The VM migration strategy of task unloading expense is minimized under mobile edge calculations environment | |
| CN104679594B (en) | A kind of middleware distributed computing method | |
| CN113615137B (en) | CDN optimization platform | |
| Mazzucco et al. | Profit-aware server allocation for green internet services | |
| Antonescu et al. | Dynamic SLA management with forecasting using multi-objective optimization | |
| CN107992353A (en) | A kind of container dynamic migration method and system based on minimum transition amount | |
| TW201447763A (en) | System and method for controlling virtual machine | |
| CN104270421A (en) | Multi-user cloud platform task scheduling method supporting bandwidth guarantee | |
| CN103970879B (en) | Method and system for regulating storage positions of data blocks | |
| CN109800059A (en) | A kind of cloud computing virtual machine migration method based on load curve similarity | |
| CN102567077A (en) | Virtualized resource distribution method based on game theory | |
| CN103973780B (en) | A kind of resource allocation of cloud video and expense optimization method | |
| CN103248622B (en) | A kind of Online Video QoS guarantee method of automatic telescopic and system | |
| CN109976879A (en) | A kind of cloud computing virtual machine placement method using curve complementation based on resource | |
| CN113656361A (en) | High-reliability data storage method and device for super-fusion power data center | |
| Kim et al. | Virtual machines placement for network isolation in clouds | |
| CN110109758A (en) | A kind of cloud computing resources distribution method | |
| CN108123893A (en) | A kind of multiple target bandwidth allocation methods of real time virtual machine migration | |
| Gilesh et al. | Towards a complete virtual data center embedding algorithm using hybrid strategy | |
| CN111598390B (en) | Method, device, equipment and readable storage medium for evaluating high availability of server | |
| Lee et al. | Resource reallocation of virtual machine in cloud computing with MCDM algorithm | |
| CN107241442A (en) | A kind of key assignments data storage storehouse copy selection method based on prediction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |