CN103729230B - Method and computer system for memory management on virtual machine system - Google Patents
Method and computer system for memory management on virtual machine system Download PDFInfo
- Publication number
- CN103729230B CN103729230B CN201310456258.8A CN201310456258A CN103729230B CN 103729230 B CN103729230 B CN 103729230B CN 201310456258 A CN201310456258 A CN 201310456258A CN 103729230 B CN103729230 B CN 103729230B
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- working set
- set size
- stored memory
- processor
- 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
Abstract
A method and a computer system for memory management on a virtual machine system are provided. The memory management method includes the following steps. First, a working set size of each of a plurality of virtual machines on the virtual machine system is obtained by at least one processor, wherein the working set size is an amount of memory required to run applications on each of the virtual machines. Then, an amount of storage memory is allocated to each of the virtual machines by the at least one processor according to the working set size of each of the virtual machines and at least one swapin or refault event, wherein the storage memory is a part of memory available from the computer system.
Description
Technical field
The present invention relates to a kind of technology that memory management is carried out on dummy machine system.
Background technology
Computer virtualized is a kind of technology for being related to virtual machine creating, and the virtual machine has operating system just as one
Physical computer, and computer virtualized framework generally by supporting multiple operations simultaneously on single physical computer platform
The ability of system is being defined.For example, running the computer of Microsoft Windows System can have with (SuSE) Linux OS with keyholed back plate
Virtual machine.Main frame is that virtualized actual physics machine occurs, and virtual machine is then counted as client computer.Super manager
（supervisor）, definitely referred to as virtual machine monitor（VMM）, it is by hardware resource virtualization and by virtual hardware interface
Present to the software layer of at least one virtual machine.Super manager is carried out with traditional operating system to the hardware resource for process
Management, and run similar relative to mode used during some management functions for performing virtual machine.Virtual machine is properly termed as " visitor
Machine ", and virtual machine internal operation operating system be properly termed as " guest operating systems ".
Virtualized environment is currently limited by internal memory, it means that the physical memory of main frame is the utilization of resources in data center
Obstruction.Physical memory resource is released and coupled by internal memory virtualization from data center, and subsequently by these resource clusterings into void
Planization memory pool, the application program that the virtualization memory pool can run by guest operating systems or on guest operating systems is entered
Row is accessed.For internal memory virtualization, memory sharing is one of important topic of memory source management and use.
When there are multiple virtual machines and running on low internal memory main frame, the internal memory distribution of virtual machine will become to application program
Performance is of crucial importance.Physical memory should be distributed between virtual machine in the way of justice, and this kind of operation is defined as " internal memory
Balance ".
The simplest form of internal memory balance is, divided by virtual machine number, and to give these virtual machines by free physical memory
In each equivalent internal memory.However, this kind of mechanism does not take into account that the working set size of each in these virtual machines, wherein
Working set size is to run the amount of ram needed for the application program in these virtual machines on each.That is, this method
It is equal that implicit assumption has each in these virtual machines, is included in the application program and input run on virtual machine
Workload.
Another kind of method is that the percentage of the free physical memory being directly proportional to each working set size is distributed to into these
Each in virtual machine.Intuition is that larger amount of internal memory is given the virtual machine to memory source with larger demand.Here
Under kind of distribution, the difference between the working set size of virtual machine and its internal memory for distributing is also with the working set size of virtual machine into just
Than.This means when the Memory Allocation of virtual machine is dropped to a fixed percentage of its working set size from its working set size
When, for the virtual machine with larger working set size, any extra event punishment（For example, cache mistake again
（refault）Or change to event (swap in)）May be higher.
There is serious degraded performance due to low memory to prevent virtual machine, it may be necessary to make in same host
The performance cost (performance overhead) of each is equal in the virtual machine of operation, and mode is by using more suitable
Internal memory balancing.
The content of the invention
The embodiment of the disclosure one is related to the EMS memory management process of the dummy machine system deposited by computer system.In described
Deposit management method to comprise the steps of.First, the multiple void on the dummy machine system are obtained by least one processor
The working set size of each in plan machine, wherein the working set size is the application journey in the operation virtual machine on each
Amount of ram needed for sequence.Subsequently, changed to according to the working set size of each in the virtual machine and at least one or
Cache mistake again, by stored memory amount is distributed in the virtual machine by least one processor each, its
Described in stored memory be free memory in the computer system a part.
Another embodiment of the disclosure is related to a kind of computer system, and the computer system includes Installed System Memory and at least one
Processor.At least one processor is couple on the Installed System Memory and performs following operation, with dummy machine system
On carry out memory management.The work of each in multiple virtual machines of at least one processor acquisition on dummy machine system
Collection size, wherein the working set size is to run the amount of ram needed for the application program in these virtual machines on each.Root
Change to according to the working set size of each in these virtual machines and at least one or cache mistake again, at described at least one
Stored memory amount is further distributed to each in these virtual machines by reason device, and wherein stored memory is can use in Installed System Memory
A part for internal memory.
If in order that aforementioned and other features and advantage of the present invention below in conjunction with accompanying drawing it can be readily appreciated that describe in detail
Dry embodiment, so as to the present invention be described in further detail.
Description of the drawings
Accompanying drawing is added to help further understand the present invention, and the accompanying drawing is incorporated in this specification and constitutes this explanation
A part for book.Accompanying drawing is used to illustrate embodiments of the invention, and is used to explain the principle of the present invention together with description.So
And, the scope of the present invention that these accompanying drawings are not intended to limit appended claims is limited.
Figure 1A is the block diagram of the computer system of an embodiment of the invention.
Figure 1B is the block diagram of the dummy machine system of an embodiment of the invention.
Fig. 2 is the EMS memory management process on the dummy machine system of an embodiment of the invention.
Specific embodiment
With detailed reference to every embodiment of the present invention, each example of the embodiment will be said in the accompanying drawings
It is bright.In the accompanying drawings and the description, same or analogous part is referred to using identical reference number as far as possible.
For illustrative purposes, a processor and an Installed System Memory used in the following embodiment, but
The present invention is not limited to this.In other embodiments, it is possible to use more than one processor.
Figure 1A is the block diagram of the computer system of an embodiment of the invention.With reference to Fig. 1, computer system 100
Comprising processor 110, Installed System Memory 120 and other standards peripheral components（It is not shown）, wherein Installed System Memory 120 couples everywhere
On reason device 110.
Processor 110 can be special or special processor, and it is configured to perform machine-readable software generation
Realizing particular task, the software code language limits the function relevant with operation to code language, so as to by with computer system
100 miscellaneous part communicates with implementing feature operation.
The softwares such as the storage program area of Installed System Memory 120, and temporarily store the currently active or commonly used data or
Application program.Therefore, Installed System Memory 120, also referred to as physical memory, it can be that a kind of access time is faster interior faster
Deposit, for example, random access memory（RAM）, static RAM（SRAM）Or dynamic random access memory
（DRAM）.
Virtual memory is a kind of technology for being managed to the resource of Installed System Memory 120.The technology is provided to be had greatly
The illusion of amount internal memory.Virtual memory and Installed System Memory 120 are divided into multiple adjacent memory address blocks, these memory address blocks
Also referred to as memory pages (page).
Super manager is arranged in computer system 100 and supports virtual machine execution space, performs in the virtual machine
In space, can change and perform multiple virtual machines by physics (entity) simultaneously.Figure 1B is the virtual of an embodiment of the invention
The block diagram of machine system.
Figure 1B is referred to together with Figure 1A, dummy machine system 100' includes multiple virtual machine 1501-150N, super manager
160 and virtual hardware 170.It should be noted that every embodiment of the present invention cover being capable of simultaneously hosted virtual machine 1501-150N
Computer system 100, and for the purpose of succinct and ease of explanation, unless otherwise stated, illustrate in the following embodiment
Two virtual machines 1501 and 150N.Each in virtual machine 1501 and 150N includes guest operating systems, for example, passenger plane behaviour
Make system 1551 or 155N, and various guest software application programs（It is not shown）.Each in guest operating systems is wrapped
Kernel containing passenger plane, for example, passenger plane kernel 1561 or 156N.Virtual hardware 170 comprising processor, internal memory and I/O devices is carried
Take and be assigned as virtual processor, virtual memory and the virtual i/o device for being connected to top operation virtual machine 1501 and 150N.Super pipe
Reason device 160 is managed to virtual machine 1501 and 150N, and provides simulation hardware and firmware resource.The one of multinomial embodiment
In individual, Linux distributions can be installed as guest operating systems 1551 and 155N in virtual machine 150, to perform any support
Application program, and support that the open source software Xen of most of Linux distribution may be provided in super manager 160.In passenger plane
Can be each each to include in dom0 kernels, and guest operating systems 1551 and 155N in core 1561 and 156N
Balloon driver（balloon driver）（It is not shown）.With reference to super manager 160, balloon driver can be by calling internal memory
Management algorithm to the virtual memory of virtual machine 1551 and 155N being allocated or deallocate.In order to reach this purpose, can be with
Changing to passenger plane kernel 1561 and 156N at and cache mistake again are intercepted, so as to quantify guest virtual machine 1501 and 150N
Performance cost, and distribute to the amount of ram of virtual machine 1501 and 150N and can be adjusted, to be by using passenger plane operation
Page regeneration (pagereclaiming) mechanism of system 1551 and 155N, makes the expense phase of each in virtual machine 1501 and 150N
Deng.
For page regeneration, processor 110 uses least recently used（Least Recently Used:LRU）Standard is come
It is determined that removing the page and keeping the order used by LRU list 157, the LRU list is according to two kinds of major type of internal memories（It is anonymous
Internal memory and page cache）The last access time, come to once being carried out by virtual machine 1501 and 150N accessed memory pages
Sequence.The memory pages of anonymous internal memory are used by the heap and stack of consumer process, and the memory pages of page cache pass through disk
Data are backed up, wherein, in first time access to after data in magnetic disk, content is buffered in internal memory to reduce following disk
I/O。
On dummy machine system 100', if the memory pages on LRU list are anonymous internal memories, then passenger plane kernel 1561
Can be by content exchange to exchange disk (swap space) with 156N（It is not shown）, to the corresponding PTE not presented in process
It is marked, subsequently discharges corresponding memory pages.Subsequently, if memory pages are accessed again, then will be by with lower section
Formula is performing COA（copy-on-access:Replicate access）Mechanism：Content of pages is brought into newly assigned internal memory from disk is exchanged
In the page, i.e. change to.Or, if the memory pages on LRU list belong to page cache, then when content of pages it is dirty
During dye, the content of pages can be refreshed output by passenger plane kernel 1561 and 156N（flush）To in exchange disk, the subsequent page
Face is released.After next file access, passenger plane kernel 1561 and 156N must again perform disk access（Referred to as again
Read）So that content is brought back to into the newly assigned page in page cache.
Fig. 2 is the EMS memory management process on the dummy machine system of an embodiment of the invention.
Before with reference to Fig. 2, it should be noted that a part for Installed System Memory 120 can be used for virtual machine 1501-150N, and be
This part of system internal memory is defined as " stored memory（storage memory）”.Referring now to Fig. 2 together with Figure 1A and Figure 1B
Part, by processor 110 obtain dummy machine system 100' on these virtual machines 1501-150N in the working set of each
Size（Step S201）.Subsequently, changed according to the working set size of each in these virtual machines 1501-150N and at least one
Enter or cache mistake again, each for by processor 110 stored memory amount being distributed in these virtual machines 1501-150N
（Step S203）.It should be noted that the summation for changing to counting and again reading off counting can be defined as expense counting, the expense is counted
It is also the number of pages being read in virtual machine.
More specifically, in of every embodiment, assume initially that each changes to the expense with cache mistake again
All it is identical on different virtual machine 1501-150N, wherein there is the different virtual machine different operating to load and different
Storage allocation amount.Given stored memory amount MAvailWith N number of virtual machine, wherein virtual machine 150i has working set size WSSi, its
Middle i=1,2 ..., N, the working set size summation of each from these virtual machines of processor 110In deduct storage
Internal memory MAvail, and subsequently subtracting each other result divided by virtual machine number N.Division result is defined as the first reduction item.Subsequently, locate
Reason device 110 is from working set size WSSiIn deduct the first reduction item, finally generate Memory Allocation or balloon mesh for virtual machine 150i
Mark (balloon target).That is, distributing to the stored memory amount of each in virtual machine 1501-150N meets equation
（1）：
Equation（1）
Wherein BTiVirtual machine 150i is distributed in expression（Or the balloon target of virtual machine 150i）Stored memory amount, WSSiTable
Show the working set size of virtual machine 150i, N represents virtual machine number, and MAvailRepresent stored memory amount.
As several examples of the present embodiment, it is assumed that two virtual machines 1501 and 150N give respectively working set size
600MB and 300MB, and assume that stored memory is 600MB.First, by the stored memory amount for lacking divided by 2（That is, 150MB）,
And identical wantage is distributed to by two virtual machines 1501 and 150N by processor 110.Subsequently, the He of virtual machine 1501
150N can be respectively allocated 450MB（Lack 150MB）And 150MB（Lack 150MB）.After Memory Allocation, virtual machine
1501 and 150N are expected to be counted with identical expense.
On different virtual machine 1501-150N identical each change to and again cache mistake expense assume not always
Correctly, because each is changed to or again the time cost of cache mistake can be with a virtual machine or between different virtual machine
It is different.For example, when the amount of ram that virtual machine is assigned to is well below its working set,（For example, to the unit of switching subsystem
Data have more modifications）, each change operation can be slowed down.In order to process such case, in another embodiment, its target
It is intended to changing between all of virtual machine 1501-150N and again reads off the time and be balanced.Each virtual machine 150i's
The summation for changing to and again reading off the time is referred to as overhead time overhead_timei, wherein i=1,2 ..., n.Virtual machine 150i
The reduction of Memory Allocation be defined as the second reduction item Si, its with change to and again read off the spent average overhead time
overhead_timeiInverse is directly proportional, because overhead time overhead_timeiIt is bigger, the Memory Allocation of virtual machine 150i
Reduce less.Subsequently, processor 110 deducts the second reduction item from working set size WSSi, final for virtual machine 150i lifes
Into Memory Allocation or balloon target.That is, distributing to the amount of ram of each in virtual machine 1501-150N meets equation
（2）：
Equation（2）
Wherein, BTiThe stored memory amount of virtual machine i, WSS are distributed in expressioniRepresent the working set size of virtual machine i, N tables
Show virtual machine number, MAvailRepresent stored memory amount.
As the numerical example of the present embodiment, it is assumed that two virtual machines 1501 and 150N also give respectively working set size 600MB
And 300MB, and stored memory is 600MB.Assume the overhead time overhead_time of virtual machine 1501 and 150NiThan for
2:1, wherein i=1, N.In the present embodiment, wantage with overhead time overhead_timeiWhat is be inversely proportional to is assigned to void
Plan machine 1501 and 150N（That is, virtual machine 1501 lacks 100MB, and virtual machine 150N lacks 200MB）.Therefore, for virtual machine
1501 final storage is assigned as 500MB, and the final storage for virtual machine 105N is assigned as 100MB.The He of virtual machine 1501
The overhead time overhead_time of 105NiExpection is balanced, and performance cost can realize equalization.
It should be noted that for the urgent memory pool of super manager 160, when processor 110 detects super manager 160
Freememory for example, during less than 1%, starts internal memory balanced controls less than the lower limit of configuration.Processor 110 can be from virtual machine
Receive in the balloon driver of each in 1501-150N and change to and again read off information, determine every in virtual machine 1501-150N
The working set of one, and correspondingly calculate new balloon target.By aforementioned EMS memory management process, all virtual machine 1501-
150N can moderately degrade, and while memory source will not excessively be lacked.
In one embodiment, can be by performing ready program on the computer of personal computer and work station etc.
To implement above-mentioned EMS memory management process.Described program is stored on computer-readable recording medium（For example, hard disk, floppy disk, CD-
ROM, MO and DVD）, read from computer readable medium, and be computer-executed.Described program can pass through internet
It is distributed Deng network.
In a word, the EMS memory management process Jing by using the existing page regeneration sector of guest operating systems, in the present invention
Design, to be changed to according to the working set size of each in these virtual machines and at least one or cache mistake again, will store
Amount of ram distributes to each virtual machine in host computer system.In this way, can make what is run in same host system
The performance cost of each is equal in virtual machine, so as to prevent virtual machine from serious performance degradation is produced due to low memory.
Those skilled in the art become readily apparent that, without departing from the scope or spirit of the invention, can be to institute
Various modifications and variations can be made for the structure of embodiments of the disclosure.In view of above, as long as the modifications and variations of the present invention belong to described
In claims and its equivalency range, then the present invention just covers the modifications and variations.
Claims (12)
1. a kind of EMS memory management process of dummy machine system, the EMS memory management process includes：
By the working set size of each in multiple virtual machines of at least one processor acquisition on the dummy machine system,
The working set size is the amount of ram needed for the application program in the operation virtual machine on each；And
Changed to according to the working set size of each in the virtual machine and at least one or cache mistake again, passed through
At least one processor is distributed to each in the virtual machine by stored memory amount, and the stored memory amount is to calculate
A part for free memory in machine system, this step also includes：
Item or the second reduction item are reduced according to the working set size of each in the virtual machine and first, by described
At least one processor is distributed to each in the virtual machine by the stored memory amount, wherein, described first reduces item
With described in each in the virtual machine number and the virtual machine on the stored memory amount, the dummy machine system
The summation of working set size is associated, wherein, described second reduces each in item and the stored memory amount, the virtual machine
The working set size summation and change to according to described at least one or again cache mistake overhead time be associated.
2. EMS memory management process according to claim 1, it is characterised in that according to each in the virtual machine
Working set size and the first reduction item, are distributed to the stored memory amount by least one processor described
The step for each in virtual machine, includes：
According to the working set size of each in each the working set size, the virtual machine in the virtual machine
Summation, and the virtual machine number on the dummy machine system, by least one processor to calculate
State the first reduction item；And
The working set size is deducted described first by least one processor and reduces item so as to by the storage
Storage distributes to each in the virtual machine.
3. EMS memory management process according to claim 2, it is characterised in that distribute to the institute of each in the virtual machine
State amount of ram and meet equation (1)：
Wherein, BTiThe stored memory amount of the virtual machine i, WSS are distributed in expressioniRepresent the work of the virtual machine i
Make collection size, N represents the virtual machine number, and MAvailRepresent the stored memory amount.
4. EMS memory management process according to claim 1, it is characterised in that according to each in the virtual machine
Working set size and the second reduction item, are distributed to the stored memory amount by least one processor described
The step for each in virtual machine, includes：
According to the working set size of each in each the working set size, the virtual machine in the virtual machine
Summation, the stored memory amount, the virtual machine number on the dummy machine system, and change according to described at least one
Enter or overhead time of cache mistake again, described second is calculated by least one processor and reduces item；
And
The working set size is deducted described second by least one processor and reduces item so as to by the storage
Storage distributes to each in the virtual machine.
5. EMS memory management process according to claim 4, it is characterised in that described second reduces item and the overhead time
It is inversely proportional to.
6. EMS memory management process according to claim 5, it is characterised in that distribute to the institute of each in the virtual machine
State stored memory amount and meet equation (2)：
Wherein, BTiThe stored memory amount of the virtual machine i, WSS are distributed in expressioniRepresent the work of the virtual machine i
Make collection size, N represents the virtual machine number, MAvailRepresent the stored memory amount, overhead_timeiRepresent virtual machine i
Overhead time.
7. a kind of computer system, it includes：
Installed System Memory；
At least one processor, at least one processor is couple on the Installed System Memory, at least one processor
Operation is performed, for carrying out memory management on dummy machine system, the operation includes：
The working set size of each in the multiple virtual machines on the dummy machine system is obtained, the working set size is fortune
The amount of ram needed for application program in the row virtual machine on each；And
Changed to according to the working set size of each in the virtual machine and at least one or cache mistake again, will be deposited
Storage amount of ram distributes to each in the virtual machine, and the stored memory amount is one of free memory in the Installed System Memory
Part,
At least one processor according to the working set size of each in the virtual machine and first reduce item or
Second reduces item, and the stored memory amount is distributed to into each in the virtual machine, wherein, described first reduces item and institute
State the work of each in the virtual machine number and the virtual machine on stored memory amount, the dummy machine system
The summation of collection size is associated, wherein, described second reduces item and the institute of each in the stored memory amount, the virtual machine
State the summation of working set size and change to according to described at least one or again cache mistake overhead time be associated.
8. computer system according to claim 7, it is characterised in that according to the work of each in the virtual machine
Make the void in the summation and the dummy machine system of the working set size of each in collection size, the virtual machine
Plan machine number, at least one processor calculates described first and reduces item, and the working set size is deducted described
First reduction item so as to the stored memory amount is distributed in the virtual machine each.
9. computer system according to claim 8, it is characterised in that distribute in the virtual machine the described of each
Amount of ram meets equation (1)：
Wherein, BTiThe stored memory amount of the virtual machine i, WSS are distributed in expressioniRepresent the work of the virtual machine i
Make collection size, N represents the virtual machine number, and MAvailRepresent the stored memory amount.
10. computer system according to claim 7, according to the working set size of each in the virtual machine,
Institute in the virtual machine on the summation of the working set size of each, the stored memory amount, the dummy machine system
State virtual machine number and change to according to described at least one or overhead time of cache mistake again, described at least one
Processor calculates described second and reduces item, and the working set is deducted described second by least one processor
Reduce item so as to the stored memory amount is distributed in the virtual machine each.
11. computer systems according to claim 10, it is characterised in that described second reduces item and the overhead time
It is inversely proportional to.
12. computer systems according to claim 11, it is characterised in that distribute to the institute of each in the virtual machine
State stored memory amount and meet equation (2)：
Wherein, BTiThe stored memory amount of the virtual machine i, WSS are distributed in expressioniRepresent the work of the virtual machine i
Make collection size, N represents the virtual machine number, MAvailRepresent the stored memory amount, overhead_timeiRepresent virtual machine i
Overhead time.
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201261712279P | 2012-10-11 | 2012-10-11 | |
| US61/712,279 | 2012-10-11 | ||
| US13/951,475 | 2013-07-26 | ||
| US13/951,475 US9128843B2 (en) | 2012-10-11 | 2013-07-26 | Method and computer system for memory management on virtual machine system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103729230A CN103729230A (en) | 2014-04-16 |
| CN103729230B true CN103729230B (en) | 2017-04-12 |
Family
ID=50453313
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310456258.8A Active CN103729230B (en) | 2012-10-11 | 2013-09-29 | Method and computer system for memory management on virtual machine system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103729230B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11176054B2 (en) | 2019-03-08 | 2021-11-16 | International Business Machines Corporation | Host virtual address space for secure interface control storage |
| US11182192B2 (en) | 2019-03-08 | 2021-11-23 | International Business Machines Corporation | Controlling access to secure storage of a virtual machine |
| US11068310B2 (en) * | 2019-03-08 | 2021-07-20 | International Business Machines Corporation | Secure storage query and donation |
| US11283800B2 (en) | 2019-03-08 | 2022-03-22 | International Business Machines Corporation | Secure interface control secure storage hardware tagging |
| US11455398B2 (en) | 2019-03-08 | 2022-09-27 | International Business Machines Corporation | Testing storage protection hardware in a secure virtual machine environment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1696902A (en) * | 2004-05-11 | 2005-11-16 | 国际商业机器公司 | System, method and program to migrate a virtual machine |
| CN101681268A (en) * | 2007-06-27 | 2010-03-24 | 国际商业机器公司 | System, method and program to manage memory of a virtual machine |
| CN101924693A (en) * | 2009-04-01 | 2010-12-22 | 威睿公司 | Be used for method and system in migrating processes between virtual machines |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7421533B2 (en) * | 2004-04-19 | 2008-09-02 | Intel Corporation | Method to manage memory in a platform with virtual machines |
-
2013
- 2013-09-29 CN CN201310456258.8A patent/CN103729230B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1696902A (en) * | 2004-05-11 | 2005-11-16 | 国际商业机器公司 | System, method and program to migrate a virtual machine |
| CN101681268A (en) * | 2007-06-27 | 2010-03-24 | 国际商业机器公司 | System, method and program to manage memory of a virtual machine |
| CN101924693A (en) * | 2009-04-01 | 2010-12-22 | 威睿公司 | Be used for method and system in migrating processes between virtual machines |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103729230A (en) | 2014-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9128843B2 (en) | Method and computer system for memory management on virtual machine system | |
| CN103729230B (en) | Method and computer system for memory management on virtual machine system | |
| Liu et al. | Hotplug or ballooning: A comparative study on dynamic memory management techniques for virtual machines | |
| CN100527098C (en) | Dynamic EMS memory mappings method of virtual machine manager | |
| KR101137172B1 (en) | System, method and program to manage memory of a virtual machine | |
| Li et al. | Composable architecture for rack scale big data computing | |
| Zhou et al. | A bare-metal and asymmetric partitioning approach to client virtualization | |
| CN102779074B (en) | Internal memory resource distribution method based on internal memory hole mechanism | |
| EP4239473A1 (en) | Container-based application management method and apparatus | |
| EP2581828B1 (en) | Method for creating virtual machine, virtual machine monitor and virtual machine system | |
| EP3230873B1 (en) | Computing method and apparatus with persistent memory | |
| EP3195128B1 (en) | Memory management in virtualized environment | |
| KR20070100367A (en) | Method, apparatus and system for dynamically reassigning memory from one virtual machine to another | |
| CN104111897A (en) | Data processing method, data processing device and computer system | |
| CN104346284B (en) | A kind of EMS memory management process and memory management equipment | |
| Song et al. | Architecting flash-based solid-state drive for high-performance I/O virtualization | |
| CN110955495B (en) | Management method, device and storage medium of virtualized memory | |
| JP2022516486A (en) | Resource management methods and equipment, electronic devices, and recording media | |
| CN106844008A (en) | A kind of method of data manipulation, equipment and system | |
| CN104102460A (en) | Cloud computing-based memory management method and device | |
| CN106445691A (en) | Memory optimization method oriented to virtual cloud computing platform | |
| CN109766179B (en) | Video memory allocation method and device | |
| Nguyen et al. | YOLO: Speeding up VM and Docker Boot Time by reducing I/O operations | |
| Chang et al. | Assessment of hypervisor and shared storage for cloud computing server | |
| US20200201691A1 (en) | Enhanced message control banks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |