CN108073443B - Virtual machine selection and placement method based on shared memory page in cloud data center - Google Patents
Virtual machine selection and placement method based on shared memory page in cloud data center Download PDFInfo
- Publication number
- CN108073443B CN108073443B CN201711294586.7A CN201711294586A CN108073443B CN 108073443 B CN108073443 B CN 108073443B CN 201711294586 A CN201711294586 A CN 201711294586A CN 108073443 B CN108073443 B CN 108073443B
- Authority
- CN
- China
- Prior art keywords
- physical
- machine
- virtual machine
- list
- virtual
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45583—Memory management, e.g. access or allocation
-
- 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
Abstract
The invention discloses a virtual machine selection and placement method based on shared memory pages in a cloud data center. According to the invention, the virtual machine selection and the virtual machine placement process are completed simultaneously, so that the transmission flow in the virtual machine heat migration process is reduced, and the energy consumption is also reduced, thereby improving the utilization rate of physical computing resources of the cloud data center.
Description
Technical Field
The invention relates to a method for selecting and placing virtual machines during live migration of the virtual machines in a cloud data center.
Background
At present, cloud computing gradually goes out of a stage of emerging technology, and more users are in the arms of cloud computing services. With the large-scale increase of the number of users, the size of each cloud data center is also expanded by adding servers. How to effectively utilize a large-scale physical server is a big problem faced by a cloud data center. Through the virtual machine live migration, the cloud data center can transfer the virtual machine from the physical machine with the physical resources being excessively used to the physical machine with the physical resources not being fully used without interrupting the service. Through the migration of the virtual machines, the load of the cloud data center is balanced, and physical resources can be more effectively utilized.
There are several challenges faced in optimizing virtual machine live migration, mainly four as follows. Firstly, physical machine overload detection: and judging which physical machines are in the CPU overload state, namely the state that the CPU resources of the physical machines are excessively used, by setting an overload threshold value. Then, detecting the light load of the physical machine: the method judges which physical machines are in the CPU light load state by setting the light load threshold, namely the state that the CPU resources of the physical machines are not fully and effectively utilized. Again, the virtual machine selection problem: how to select the virtual machine from the physical machine in the overload state so that the physical machine is no longer in the overload state. The last is the virtual machine placement problem: by solving the virtual machine selection problem, a group of virtual machines to be migrated is obtained, and appropriate physical machines which are not overloaded are selected to place the virtual machines. In general, the virtual machine selection problem and the virtual machine placement problem are solved in a sequential order.
Various virtual machine selection strategies have been proposed. A minimum number of migrations (MM) policy selects the least number of virtual machines to migrate so that the overloaded physical machine is no longer in an overloaded state. A potential growth (HPG) policy selects those virtual machines migration that have minimal CPU resource usage, thereby reducing the load on the physical machine. In a random selection (RC/RS) strategy, virtual machines are randomly selected to migrate until the physical machine is no longer overloaded. In a Minimum Migration Time (MMT) policy, those virtual machines with the least amount of memory usage will be migrated given the network bandwidth. The maximum correlation policy (MC) policy selects those virtual machines with the greatest CPU correlation for migration. The minimum CPU usage (MinU) policy selects the virtual machine occupying the minimum CPU resource amount for migration, and the maximum CPU usage (MaxU) policy selects the virtual machine occupying the maximum CPU resource amount for migration. Experiments in reference [ 1 ] show that, among the above virtual machine selection strategies, the MMT strategy is superior to other virtual machine selection strategies in performance indexes such as energy consumption and Service Level Agreement Violation (SLAV). The reason for this is that using the MMT policy results in the least amount of memory of the virtual machine that needs to be transmitted, while other virtual machine selection policies only consider the usage of CPU resources by the virtual machine.
Research finds that great similarity exists between the memory contents of the virtual machines. The memory of the virtual machine is stored by taking a memory page as a basic unit. From the perspective of the memory pages, it is observed that many memory pages are identical or similar between virtual machines using the same or similar operating systems and the same application software. By utilizing the property, the memory amount of the virtual machine, which needs to be transmitted in the process of virtual machine live migration, can be greatly reduced: when a plurality of virtual machines are migrated to one physical machine, the same memory page is only required to be transmitted once.
Therefore, it is necessary to design a virtual machine selection and placement method that can minimize the amount of memory of the virtual machine that needs to be transmitted and can further optimize performance indexes such as energy consumption and SLAV.
Disclosure of Invention
The invention aims to solve the technical problem that the virtual machine selection and placement method based on the shared memory page in the cloud data center is provided for overcoming the defects of the prior art, so that the network pressure, the SLAV and the energy consumption of the cloud data center during the thermal migration of the virtual machine can be optimized.
The technical scheme provided by the invention for solving the technical problems is as follows:
a virtual machine selection and placement method based on a shared memory page in a cloud data center comprises the following steps:
the method comprises the following steps: inputting a list of m physical machines which are not overloaded, a list of s physical machines which are overloaded and a list of virtual machines which run on each physical machine, and recording the m physical machines which are not overloaded as pm1～pmmThe list of overloaded s physical machines is pmm+1～pmm+sAll overloaded physical machines runningThe set formed by the virtual machines is V;
step two: arranging non-overloaded non-empty physical machines in a descending order according to the available resource quantity of the CPU, and then adding unloaded physical machines behind the non-overloaded non-empty physical machines to form an initial physical machine List PM _ List; by using
Representing a set formed by all virtual machines ready to be migrated to the tth physical machine in the initial physical machine List PM _ List in the ith round of iteration, wherein t is 1,2, …, m; initialization
t＝1；
Step two: the first physical machine in the current physical machine List PM _ List is denoted as PM1, and the set of memory pages included in the virtual machine determined on PM1 is denoted as p1The set of all virtual machines on the PM1 is denoted as VM, and the overload threshold of the PM1 is denoted as t (VM); initializing an iteration counter i to be 0;
step four: for any virtual machine vm in V, order pi(vm)＝p(vm)\p1Wherein, the symbol \ represents the difference set between the two sets, p (vm) is the memory page set contained in the vm of the virtual machine, pi(vm) is a memory page set which needs to be transmitted by the virtual machine vm in the ith iteration;
step five: determine if any j ∈ [ m +1, m + s)]All have Tj(Vj) Not equal to 1, if yes, order
The algorithm ends, where VmtIs a virtual machine set which is obtained at the end of the algorithm and is migrated to the t-th physical machine in the initial physical machine List PM _ List, wherein VjIs a physical machine pmjSet of running virtual machines, Tj(Vj) Is a physical machine pmjThe overload threshold value is not equal to 1, which indicates that the physical machine is not overloaded, and the overload threshold value is equal to 1, which indicates that the physical machine is overloaded; otherwise, entering the step six;
step six: determining whether there is T (VM) equal to 1,if yes, indicating that PM1 is already in an overload state, then command
For any virtual machine vm in V, let p (vm) be pi(vm)∪p1Deleting PM1 from the PM _ List, making t equal to t +1, and then jumping to the third step; otherwise, entering the step seven;
step seven: judging whether j belongs to [ m +1, m + s ] or not]There is a physical machine pmjHaving a Tj(Vj) Not equal to 1, the physical machine is switched from the overload state to the non-overload state, if yes, V is set as V \ VjEntering the step eight; otherwise, directly entering the step eight;
step eight: selecting a virtual machine from V, and recording the virtual machine selected in the ith iteration as vmiWhich satisfies
To minimize, mark the overloaded physical machine running the virtual machine as pmqWherein p isi(vmi) For the virtual machine vm in the ith iterationiMemory page set, | p, to be transmittedi(vmi) L is pi(vmi) Number of memory pages in c (vm)i) For virtual machines vmiThe amount of CPU resources used;
step nine: virtual machine vmiPlacing the VM (virtual machine) on a PM1 (physical machine) to be VM ═ VM ≈ VMi， p1＝p1∪pi(vmi)，
Vq＝Vq\vmi，VqIs a physical machine pmqA set of running virtual machines; for any virtual machine vm in V, order pi+1(vm)＝pi(vm)\pi(vmi) (ii) a And i is made to be i +1, and the step five is skipped.
In the first step, the input list of m physical machines which are not overloaded and the list of s physical machines which are overloaded comprise each physical machine pmjIs an overload threshold value Tj(Vj) Amount of resources available to CPU CA at idlej，j∈[1，m+s]In which V isjIs a physical machine pmjAnd the set formed by the upper running virtual machines is obtained by inputting a virtual machine list of each physical machine running. The input list of virtual machines run by each physical machine should also include the memory page set and the amount of CPU resources used by each virtual machine.
The overload threshold T in the fifth stepj(Vj) From the overload threshold function FT, IQR, MAD [ IQR (V)i)、 MAD(Vi) The calculation method is described in detail in reference [ 1 ] and optionally converted into the following specific conversion method.
For a fixed threshold FT (FT (V)j)＝0.81·CAj) Comprises the following steps:
for IQR there are:
for MAD there are:
in the second step, the calculation formula of the available resource amount of the CPU of the non-empty physical machine is as follows:
wherein, CjIs a physical machine pmjAmount of CPU available resources, CAjIs a physical machine pmjAmount of CPU available resources, V, at idlejIs a physical machine pmjSet of running virtual machinesAnd c (vm) is the amount of CPU resources used by the virtual machine vm.
The invention greatly reduces the memory amount, SLA and energy consumption of the virtual machine which needs to be transmitted during the hot migration based on the following principles:
the memory of the virtual machine is stored in the memory of the physical machine by taking the memory page as a basic unit. There is a great similarity between the memory contents of different virtual machines, so many memory pages are the same between them. When the virtual machines are migrated in a live mode, if multiple virtual machines are migrated to one physical machine, the same memory page of the virtual machine only needs to be transmitted once.
Has the advantages that:
the invention can realize that the overload physical machine is converted into the non-overload state, when the virtual machine is selected to be migrated and placed on the overload physical machine, the number of the memory pages of the virtual machine to be transmitted is minimized under the condition of considering the use of CPU resources, and the virtual machine with the number of the memory pages to be transmitted as small as possible is selected each time until the physical machine is not in the overload state any more. The method can further optimize the MMT strategy, solves the virtual machine selection problem and the virtual machine placement problem by combining the two problems, and greatly reduces the memory amount, SLA and energy consumption of the virtual machines which need to be transmitted during thermal migration, thereby better optimizing the resource allocation of the cloud data center, reducing the network pressure and improving the utilization rate of physical computing resources.
Drawings
FIG. 1 is a flow chart of the present invention.
Fig. 2 shows a comparison of the individual algorithms with respect to the SLAV performance indicators.
FIG. 3 is a comparison of the number of virtual machine memory pages required to be transferred for each algorithm.
Fig. 4 is a comparison of the performance indicators of the energy consumption for each algorithm.
In fig. 2 to 4, the algorithm proposed by the present invention is CVSP.
Detailed Description
The invention will be further described with reference to the accompanying drawings.
The invention discloses a virtual machine selection and placement method based on a shared memory page in a cloud data center, which comprises the following steps of:
the method comprises the following steps: inputting a list of m physical machines which are not overloaded, a list of s physical machines which are overloaded and a list of virtual machines which run on each physical machine, and recording the m physical machines which are not overloaded as pm1～pmmThe list of overloaded s physical machines is pmm+1～pmm+sThe set formed by the virtual machines running on all overloaded physical machines is V;
step two: arranging non-overloaded non-empty physical machines in a descending order according to the available resource quantity of the CPU, and then adding unloaded physical machines behind the non-overloaded non-empty physical machines to form an initial physical machine List PM _ List; by using
Representing a set formed by all virtual machines ready to be migrated to the tth physical machine in the initial physical machine List PM _ List in the ith round of iteration, wherein t is 1,2, …, m; initialization
t＝1；
Step three: the first physical machine in the current physical machine List PM _ List is denoted as PM1, and the set of memory pages included in all virtual machines on PM1 is denoted as p1The set of all virtual machines on the PM1 is denoted as VM, and the overload threshold of the PM1 is denoted as t (VM); initializing an iteration counter i to be 0;
step four: for any virtual machine vm in V, order pi(vm)＝p(vm)\p1Wherein, the symbol \ represents the difference set between the two sets, p (vm) is the memory page set contained in the vm of the virtual machine, pi(vm) is a memory page set which needs to be transmitted by the virtual machine vm in the ith iteration;
step five: determine if any j ∈ [ m +1, m + s)]All have Tj(Vj) Not equal to 1, if yes, order
The algorithm ends, where VmtFor migration to the initial physical machine list PM _ Li obtained at the end of the algorithmset of virtual machines on the t-th physical machine in st, where VjIs a physical machine pmjSet of running virtual machines, Tj(Vj) Is a physical machine pmjThe overload threshold value is not equal to 1, which indicates that the physical machine is not overloaded, and the overload threshold value is equal to 1, which indicates that the physical machine is overloaded; otherwise, entering the step six;
step six: judging whether T (VM) is equal to 1, if so, indicating that PM1 is in overload state, and controlling
For any virtual machine vm in V, let p (vm) be pi(vm)∪p1Deleting PM1 from the PM _ List, making t equal to t +1, and then jumping to the third step; otherwise, entering the step seven;
step seven: judging whether j belongs to [ m +1, m + s ] or not]There is a physical machine pmjHaving a Tj(Vj) Not equal to 1, the physical machine is switched from the overload state to the non-overload state, if yes, V is set as V \ VjEntering the step eight; otherwise, directly entering the step eight;
step eight: selecting a virtual machine from all overloaded physical machines, and recording the virtual machine selected in the ith iteration as vmi which satisfies the condition
To minimize, mark the overloaded physical machine running the virtual machine as pmqWherein p isi(vmi) For the virtual machine vm in the ith iterationiMemory page set, | p, to be transmittedi(vmi) L is pi(vmi) Number of memory pages in c (vm)i) For virtual machines vmiThe amount of CPU resources used;
step nine: virtual machine vmiPlaced on physical machine PM1, order
p1＝p1∪pi(vmi)，
Vq＝Vq\vmi，VqIs a physical machine pmqA set of running virtual machines; for any virtual machine vm in V, order pi+1(vm)＝pi(vm)\pi(vmi) (ii) a And i is made to be i +1, and the step five is skipped.
The invention uses the plant Lab distributed cluster provided in the reference (2) to perform the performance test by using Matlab2000 as a simulation platform on the basis of the use record (once every five minutes) of approximately 4000 virtual machines to the CPU within one day. The algorithm for comparison is a virtual machine selection algorithm such as RS, MC, MMT and the like proposed in the reference document [ 1 ] to combine the PABFD virtual machine placement algorithm proposed in the reference document [ 1 ]. The threshold function used in the comparison is the FT, MAD, IQR overload threshold function proposed in reference [ 1 ].
As can be seen from fig. 2 to 4, regarding the SLAV, the number of pages of the memory of the virtual machine to be transmitted, and the energy consumption, the CVSP algorithm proposed by the present invention is superior to the combination of the PABFD and various virtual machine selection strategies under the condition of adopting different overload threshold functions.
Reference documents:
【1】A.Beloglazov,R.Buyya,Optimal online deterministic algorithms and adaptive heuristics for energy and performance efficient dynamic consolidation of virtual machines in cloud data centers,Concurrency and Computation:Practice and Experience 24(13)(2012) 1397–1420.
【2】A.Beloglazov,R.Buyya,Managing overloaded hosts for dynamic consolidation of virtual machines in cloud data centers under quality ofservice constraints,IEEE Transactions on Parallel and Distributed Systems 24(7)(2013)1366–1379.
Claims (4)
1. a virtual machine selection and placement method based on a shared memory page in a cloud data center is characterized by comprising the following steps:
the method comprises the following steps: input has not passedThe loaded m physical machines list, the overloaded s physical machines list and the virtual machine list running on each physical machine are recorded, and the m physical machines which are not overloaded are recorded as pm1～pmmThe list of overloaded s physical machines is pmm+1～pmm+sThe set formed by the virtual machines running on all overloaded physical machines is V;
step two: arranging non-overloaded non-empty physical machines in a descending order according to the available resource quantity of the CPU, and then adding unloaded physical machines behind the non-overloaded non-empty physical machines to form an initial physical machine List PM _ List; by using
Representing a set formed by all virtual machines ready to be migrated to the tth physical machine in the initial physical machine List PM _ List in the ith round of iteration, wherein t is 1,2, …, m; initialization
Step three: the first physical machine in the current physical machine List PM _ List is denoted as PM1, and the set of memory pages included in all virtual machines on PM1 is denoted as p1The set of all virtual machines on the PM1 is denoted as VM, and the overload threshold of the PM1 is denoted as t (VM); initializing an iteration counter i to be 0;
step four: for any virtual machine vm in V, order pi(vm)＝p(vm)\p1Wherein, the symbol \ represents the difference set between the two sets, p (vm) is the memory page set contained in the vm of the virtual machine, pi(vm) is a memory page set which needs to be transmitted by the virtual machine vm in the ith iteration;
step five: determine if any j ∈ [ m +1, m + s)]All have Tj(Vj) Not equal to 1, if yes, order
The algorithm ends, where VmtIs a virtual machine set which is obtained at the end of the algorithm and is migrated to the t-th physical machine in the initial physical machine List PM _ List, wherein VjIs a physical machine pmjVirtual machine running onSet of (2), Tj(Vj) Is a physical machine pmjThe overload threshold value is not equal to 1, which indicates that the physical machine is not overloaded, and the overload threshold value is equal to 1, which indicates that the physical machine is overloaded; otherwise, entering the step six;
step six: judging whether T (VM) is equal to 1, if so, indicating that PM1 is in overload state, and controlling
For any virtual machine vm in V, let p (vm) be pi(vm)∪p1Deleting PM1 from the PM _ List, making t equal to t +1, and then jumping to the third step; otherwise, entering the step seven;
step seven: judging whether j belongs to [ m +1, m + s ] or not]There is a physical machine pmjHaving a Tj(Vj) Not equal to 1, the physical machine is switched from the overload state to the non-overload state, if yes, V is set as V \ VjEntering the step eight; otherwise, directly entering the step eight;
step eight: selecting a virtual machine from V, and recording the virtual machine selected in the ith iteration as vmiWhich satisfies
To minimize, mark the overloaded physical machine running the virtual machine as pmqWherein p isi(vmi) For the virtual machine vm in the ith iterationiMemory page set, | p, to be transmittedi(vmi) L is pi(vmi) Number of memory pages in c (vm)i) For virtual machines vmiThe amount of CPU resources used;
step nine: virtual machine vmiPlacing the VM (virtual machine) on a PM1 (physical machine) to be VM ═ VM ≈ VMi ， p1＝p1∪pi(vmi)，
Vq＝Vq\vmi，VqIs a physical machine pmqVirtual machine organization of upper runA set of; for any virtual machine vm in V, order pi+1(vm)＝pi(vm)\pi(vmi) (ii) a And i is made to be i +1, and the step five is skipped.
2. The method for selecting and placing virtual machines based on shared memory pages in cloud data center according to claim 1, wherein in the first step, the input list of m physical machines which are not overloaded and the list of s physical machines which are overloaded comprise each physical machine pmjIs an overload threshold value Tj(Vj) Amount of resources available to CPU CA at idlej，j∈[1,m+s](ii) a Physical machine pmjSet V formed by virtual machines running onjObtaining a virtual machine list operated by each input physical machine; the input virtual machine list operated by each physical machine comprises the memory page set contained in each virtual machine and the used CPU resource amount.
3. The method for selecting and placing virtual machines based on shared memory pages in cloud data center according to claim 1, wherein the overload threshold T in the fifth stepj(Vj) By an overload threshold function FT (V)j) Four-bit distance IQR statistical method IQR (V)j) Or median absolute difference MAD statistical method MAD (V)j) The transformation is carried out in the following specific way:
for FT there are:
wherein FT (V)j)＝0.81·CAj，CAjIs a physical machine pmjAvailable resource amount of CPU in idle time;
for IQR there are:
for MAD there are:
4. the method for selecting and placing the virtual machine based on the shared memory page in the cloud data center according to claim 1, wherein in the second step, a calculation formula of the available resource amount of the CPU of the non-empty physical machine is as follows:
wherein, CjIs a physical machine pmjAmount of CPU available resources, CAjIs a physical machine pmjThe amount of resources available to the CPU at idle, c (vm), is the amount of CPU resources used by the virtual machine vm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711294586.7A CN108073443B (en) | 2017-12-08 | 2017-12-08 | Virtual machine selection and placement method based on shared memory page in cloud data center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711294586.7A CN108073443B (en) | 2017-12-08 | 2017-12-08 | Virtual machine selection and placement method based on shared memory page in cloud data center |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108073443A CN108073443A (en) | 2018-05-25 |
| CN108073443B true CN108073443B (en) | 2021-10-26 |
Family
ID=62158176
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711294586.7A Active CN108073443B (en) | 2017-12-08 | 2017-12-08 | Virtual machine selection and placement method based on shared memory page in cloud data center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108073443B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113157435B (en) * | 2021-03-01 | 2023-11-24 | 长沙学院 | Virtual machine deployment method and system with energy efficiency as target of cloud data center |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103524A (en) * | 2010-12-23 | 2011-06-22 | 北京航空航天大学 | Memory redundancy oriented virtual machine migration device and method |
| CN102968344A (en) * | 2012-11-26 | 2013-03-13 | 北京航空航天大学 | Method for migration scheduling of multiple virtual machines |
| CN103440158A (en) * | 2013-07-29 | 2013-12-11 | 西北工业大学 | Hotspot migration method for cloud resource scheduling |
| CN104035542A (en) * | 2014-05-28 | 2014-09-10 | 中国科学院计算技术研究所 | Virtual machine migration method and system for balancing calculation energy consumption and refrigeration energy consumption |
| CN105430083A (en) * | 2015-11-27 | 2016-03-23 | 成都微讯云通科技有限公司 | Cloud platform deployment method |
| CN106502765A (en) * | 2016-11-07 | 2017-03-15 | 中南大学 | A kind of virtual machine placement method suitable for cloud data center |
| CN106598733A (en) * | 2016-12-08 | 2017-04-26 | 南京航空航天大学 | Three-dimensional virtual resource scheduling method of cloud computing energy consumption key |
| CN106933650A (en) * | 2017-03-03 | 2017-07-07 | 北方工业大学 | load management method and system of cloud application system |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8230069B2 (en) * | 2008-03-04 | 2012-07-24 | International Business Machines Corporation | Server and storage-aware method for selecting virtual machine migration targets |
| US9489222B2 (en) * | 2011-08-24 | 2016-11-08 | Radware, Ltd. | Techniques for workload balancing among a plurality of physical machines |
| US9886301B2 (en) * | 2015-05-04 | 2018-02-06 | Strato Scale Ltd. | Probabilistic deduplication-aware workload migration |
-
2017
- 2017-12-08 CN CN201711294586.7A patent/CN108073443B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103524A (en) * | 2010-12-23 | 2011-06-22 | 北京航空航天大学 | Memory redundancy oriented virtual machine migration device and method |
| CN102968344A (en) * | 2012-11-26 | 2013-03-13 | 北京航空航天大学 | Method for migration scheduling of multiple virtual machines |
| CN103440158A (en) * | 2013-07-29 | 2013-12-11 | 西北工业大学 | Hotspot migration method for cloud resource scheduling |
| CN104035542A (en) * | 2014-05-28 | 2014-09-10 | 中国科学院计算技术研究所 | Virtual machine migration method and system for balancing calculation energy consumption and refrigeration energy consumption |
| CN105430083A (en) * | 2015-11-27 | 2016-03-23 | 成都微讯云通科技有限公司 | Cloud platform deployment method |
| CN106502765A (en) * | 2016-11-07 | 2017-03-15 | 中南大学 | A kind of virtual machine placement method suitable for cloud data center |
| CN106598733A (en) * | 2016-12-08 | 2017-04-26 | 南京航空航天大学 | Three-dimensional virtual resource scheduling method of cloud computing energy consumption key |
| CN106933650A (en) * | 2017-03-03 | 2017-07-07 | 北方工业大学 | load management method and system of cloud application system |
Non-Patent Citations (5)
| Title |
|---|
| An Energy Efficient VM Migration Algorithm in Data Centers;Xiaodong Wu 等;《2017 16th International Symposium on Distributed Computing and Applications to Business, Engineering and Science (DCABES)》;20171116;27-30 * |
| An Improved Priority Based VM Placement Approach for the Cloud Environment;Prafulla Chakankar 等;《IJCSN International Journal of Computer Science and Network》;20160430;第5卷(第2期);210-214 * |
| Energy Efficient Virtual Machine Placement Algorithm With Balanced Resource Utilization Based on Priority of Resources;Amin Rahimi 等;《Computer Engineering and Applications》;20150630;第4卷(第2期);108-118 * |
| 基于自适应过载阈值选择的虚拟机动态整合方法;闫成雨 等;《计算机应用》;20161010;第36卷(第10期);2698-2703，2709 * |
| 基于负载特征的虚拟机动态放置策略研究;高龙;《中国优秀硕士学位论文全文数据库 信息科技辑》;20141215(第12期);I137-4 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108073443A (en) | 2018-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Piao et al. | A network-aware virtual machine placement and migration approach in cloud computing | |
| Fan et al. | Workload allocation in hierarchical cloudlet networks | |
| US20190324819A1 (en) | Distributed-system task assignment method and apparatus | |
| US9442763B2 (en) | Resource allocation method and resource management platform | |
| Esfandiarpoor et al. | Structure-aware online virtual machine consolidation for datacenter energy improvement in cloud computing | |
| CN102236582B (en) | Method for balanced distribution of virtualized cluster load in a plurality of physical machines | |
| EP3281359B1 (en) | Application driven and adaptive unified resource management for data centers with multi-resource schedulable unit (mrsu) | |
| CN103294548B (en) | A kind of I/O request dispatching method based on distributed file system and system | |
| US10884667B2 (en) | Storage controller and IO request processing method | |
| CN106775949B (en) | Virtual machine online migration optimization method capable of sensing composite application characteristics and network bandwidth | |
| CN111338801B (en) | Subtree migration method and device for realizing metadata load balance | |
| WO2020134133A1 (en) | Resource allocation method, substation, and computer-readable storage medium | |
| Tao et al. | Load feedback-based resource scheduling and dynamic migration-based data locality for virtual hadoop clusters in openstack-based clouds | |
| CN109976879B (en) | Cloud computing virtual machine placement method based on resource usage curve complementation | |
| CN112954012B (en) | Cloud task scheduling method based on improved simulated annealing algorithm of load | |
| US20230379268A1 (en) | Resource scheduling method and system, electronic device, computer readable storage medium | |
| US10671293B2 (en) | Topology aware load optimized multipath I/O scheduler | |
| CN108073443B (en) | Virtual machine selection and placement method based on shared memory page in cloud data center | |
| CN110362379A (en) | Based on the dispatching method of virtual machine for improving ant group algorithm | |
| CN110515564B (en) | Method and device for determining input/output (I/O) path | |
| Wang et al. | A throughput driven task scheduler for improving mapreduce performance in job-intensive environments | |
| CN110308965B (en) | Rule-based heuristic virtual machine distribution method and system for cloud data center | |
| Ashalatha et al. | Dynamic load balancing methods for resource optimization in cloud computing environment | |
| Kumar et al. | A priority based dynamic load balancing approach in a grid based distributed computing network | |
| CN115718644A (en) | Computing task cross-region migration method and system for cloud data center |
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 |