CN107423114B - Virtual machine dynamic migration method based on service classification - Google Patents
Virtual machine dynamic migration method based on service classification Download PDFInfo
- Publication number
- CN107423114B CN107423114B CN201710582510.8A CN201710582510A CN107423114B CN 107423114 B CN107423114 B CN 107423114B CN 201710582510 A CN201710582510 A CN 201710582510A CN 107423114 B CN107423114 B CN 107423114B
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- memory
- service
- migration
- utilization rate
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Debugging And Monitoring (AREA)
Abstract
The invention provides a dynamic virtual machine migration method based on service classification, which comprises the following steps: 1) acquiring the comprehensive utilization rate of the virtual machine; 2) when the comprehensive utilization rate of the virtual machine is greater than a threshold value, evaluating the comprehensive utilization rate of each service of the virtual machine, classifying each service according to the dividing requirement, and selecting a target virtual machine according to the classification result; wherein the threshold and partitioning requirements are defined by an administrator according to different scenarios; 3) and migrating the classified related resources of the partial services to the target virtual machine. The invention considers the analysis of the use condition of the virtual machine and only migrates partial services and users. Therefore, the method of the invention not only can improve the overall migration performance, but also can improve the use efficiency of resources.
Description
Technical Field
The invention relates to a virtual machine migration method, in particular to a virtual machine dynamic migration method based on service classification.
Background
In recent years, cloud computing is widely popularized and applied, and as the use time is increased, the consumption and shortage of resources are increasingly obvious. Currently, most of the methods adopt virtual machine migration, that is, the entire Virtual Machine (VM) is migrated. However, if the resources of the virtual machine in the cloud platform are limited, sufficient resources may not be provided for the overall migration. The virtual machine, as an important component in the cloud computing environment, bears the use and operation of the user-related resources, and also becomes the core of the migration process. The virtual machine migration process is mainly divided into cold migration and hot migration, and the hot migration is also called dynamic migration. The cold migration is the whole migration after the virtual machine is stopped, but influences the use of users; the dynamic migration is to complete the migration of the virtual machine without the user's awareness, and it ensures that the virtual machine still runs smoothly during the migration process, and the user does not perceive any difference, thus becoming the focus of research and application.
The existing dynamic migration mode mainly comprises pre-migration, post-migration and the like. With premigration being the most used approach. The pre-migration mainly refers to sending the memory data of the source virtual machine to the target virtual machine through continuous circulation in the migration process. The pre-copy mechanism greatly reduces the amount of memory data to be transmitted in the shutdown copy stage, thereby greatly reducing the shutdown time. However, for a memory portion with a very fast update speed, each loop process becomes dirty, and needs to be copied repeatedly, which also results in a very large number of loops and a long migration time. The post-migration mainly comprises the steps of stopping and sending the CPU state in the source virtual machine and the minimum working state which can enable the virtual machine to recover to run to the target virtual machine, starting the target virtual machine to run, and gradually migrating the missing memory pages. However, the method is not adopted much because the running performance is seriously affected by page fault interruption in the target virtual machine, and the service performance is reduced.
In the actual migration process, the same virtual machine provides different services for a plurality of users, but when the user access amount is increased and the resources are insufficient, the resources need to be migrated. The conventional method is to migrate the whole virtual machine to a virtual machine with larger resources, but on one hand, there may not be a virtual machine meeting the requirements, and on the other hand, the complete migration results in a long time.
Disclosure of Invention
Aiming at the problems, the invention considers the analysis of the use condition of the virtual machine and only migrates part of services and users. Therefore, the invention provides a dynamic migration method of a virtual machine based on service classification, which not only can improve the overall migration performance, but also can improve the use efficiency of resources.
Referring to fig. 1, the present invention mainly performs effective division on services in a migration process based on pre-migration, migrates relatively independent resources in the services, and can effectively combine with the existing method to improve efficiency in a dynamic migration process. Where the VMM in this figure represents a virtual machine monitor.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a dynamic migration method of virtual machines based on service classification comprises the following steps:
1) acquiring the comprehensive utilization rate of the virtual machine;
2) when the comprehensive utilization rate of the virtual machine is greater than a threshold value, evaluating the comprehensive utilization rate of each service of the virtual machine, classifying each service according to the dividing requirement, and selecting a target virtual machine according to the classification result; wherein the threshold and partitioning requirements are defined by an administrator according to different scenarios;
3) and migrating the classified related resources of the partial services to the target virtual machine.
Further, before step 1), the method further comprises: monitoring the use condition of related resources of each virtual machine within a preset time through a virtual machine monitor; the usage conditions of the resources related to the virtual machine comprise the sizes of a CPU, a memory and a hard disk and the average usage rates of the CPU, the memory and the hard disk in preset time.
Further, the comprehensive utilization rate Φ ═ α · Cr + β · Mr + γ · Hr in step 1), where α, β, γ are impact weights and are defined by an administrator according to different scenarios; cr, Mr and Hr respectively refer to the average utilization rate of the CPU, the memory and the hard disk in a preset time.
Further, step 2) also includes: if no suitable target virtual machine exists, the services are reclassified.
Further, the size of the free space of the CPU, the memory and the hard disk of the target virtual machine in the step 2) is larger than sigma times of the utilization rate of the service set to be migrated; the sigma is more than or equal to 2, and the service set to be migrated is a service set to be migrated which is divided according to the dividing requirement.
Further, in step 2), when the comprehensive utilization rate of the virtual machines is greater than a threshold value, executing the following steps:
2-1) calculating the comprehensive utilization rate of each service in the virtual machine, sorting the services from large to small, and classifying the services according to a sorting result and a dividing requirement;
2-2) selecting the target virtual machine according to the classification result.
Furthermore, in the step 3), the migration is performed according to the sequence of the sorted comprehensive utilization rate of each service, and the sequence of the migration of the relevant resources of the service is determined according to the change condition of the relevant resources of the service within the monitored preset time.
Furthermore, when the related resources of the service are not changed, the migration is directly carried out; when the related resources of the service change, monitoring the changed service, sequencing according to the service change speed, and migrating according to the principle that the related resources of the service change from less to more; the related resources comprise host names (id), total CPU (Cs), CPU usage (Cu), total memory (Ms), memory usage (Mu), total hard disk (Hs) and hard disk usage (Hu).
Further, step 3) further comprises: the hard disk is set to shared storage.
Further, step 3) further comprises: intensively and rapidly migrating the memory by adopting a pre-copying method; wherein the concentrated fast migration of the pre-copy method refers to: marking the memory change part by using the shadow memory, copying the memory change part to an idle memory area, and continuously migrating until the content of the memory change part can be migrated at one time within tau time; where τ is defined by the administrator.
The invention has the beneficial effects that: the invention provides a dynamic migration method of a virtual machine based on service classification, which determines whether migration is needed or not and a target needing migration after the use condition of related resources of the virtual machine is collected. If the migration is needed, different services are classified, and a proper virtual machine is effectively selected to perform the related resource migration of the services, so that the migration efficiency is greatly improved. In the migration process, the method sets the related data of the hard disk as shared storage, so that the data sharing and storage are convenient; in the aspect of memory, the whole memory is migrated, then the memory change part is marked by means of a shadow memory, the memory change part is continuously copied to a free memory area, and then the integral migration is carried out until the last synchronization is carried out.
Drawings
Fig. 1 is a schematic diagram of related resource migration of the service provided by the present invention.
Fig. 2 is a flowchart of a dynamic migration method of a virtual machine based on service classification according to the present invention.
FIG. 3 is a diagram illustrating a marked memory change portion and a mapping of the marked memory change portion to a free memory area according to an embodiment of the invention.
Detailed Description
In order to make the aforementioned and other features and advantages of the invention more comprehensible, embodiments accompanied with figures are described in detail below.
The invention provides a dynamic migration method of a virtual machine based on service classification, please refer to fig. 2, which includes the following steps:
1. and acquiring the comprehensive utilization rate of the virtual machine.
And monitoring the use condition of each virtual machine within a preset time through a virtual machine monitor of the virtual machine monitoring platform to obtain the use condition of the related resources of the virtual machine, and obtaining the comprehensive utilization rate of the virtual machine according to the use condition of the related resources of the virtual machine. The method for acquiring the service conditions of the related resources of the virtual machine mainly comprises the steps of extracting the sizes of a CPU (Central processing Unit), a memory and a hard disk in the virtual machine and the average utilization rate of the CPU, the memory and the hard disk in a preset time, and constructing a related feature set, so that the subsequent service conditions of the resources can be conveniently judged.
2. And when the comprehensive utilization rate of the virtual machine is greater than a threshold value, evaluating the comprehensive utilization rate of each service of the virtual machine, classifying each service according to the division requirements, and selecting a target virtual machine (namely a proper idle virtual machine) according to the classification result.
Setting a corresponding threshold value for the comprehensive utilization rate of the virtual machine obtained according to the use condition of the related resources of the virtual machine, and when the threshold value is greater than the threshold value, evaluating the comprehensive utilization rate of each service of the virtual machine, and classifying each service according to the division requirement; wherein the threshold and the partitioning requirement are defined by an administrator according to different scenarios. Meanwhile, the idle virtual machine resources are evaluated, and the related resources are migrated.
3. And (5) a migration process.
And determining the sequence of the related resource migration according to the change condition of the related resources served within the monitored preset time. In the migration process, when the related resources of the service are not changed, the migration is directly carried out; when the relevant resources of the service change, the changed service needs to be monitored, and the services are sequenced according to the service change speed, and are sequentially migrated according to the principle that the relevant resources of the service change from less to more until the end. In the specific migration process, in the aspect of the hard disk, related data of the hard disk is set to be shared and stored, so that the data sharing and storage are convenient; in the aspect of memory, the whole memory is migrated, then the memory change part is marked by means of a shadow memory, the memory change part is continuously copied to a free memory area, and then the integral migration is carried out until the last synchronization is carried out.
The process of the invention is further illustrated below by way of a specific example. The steps of this embodiment include:
1. and acquiring the service condition of the related resources of the virtual machine, and further acquiring the comprehensive utilization rate of the virtual machine. Because of the virtual machine architecture, there is usually a virtual machine monitor to implement centralized management of the physical and virtual IT infrastructure, improve server utilization, and optimize dynamic resources across multiple virtualization platforms. Therefore, the method can read the usage situation E ═ ξ } of related resources of each virtual machine in the platform through the virtual machine monitor. libvirt provides an API to securely manage guest operating systems running on a host computer. It is a programming interface that implements a virtualization tool, supporting various virtual machine monitors, such as Xen, KVM, QEMU, etc. Meanwhile, libvirt can receive user instructions and call or monitor corresponding instructions and services in the virtual machine. The method mainly writes related codes to check the use condition of the related resources of the virtual machine by calling a libvirt interface and combining C and Python languages.
First, the related resources of the virtual machine mainly concerned by the present invention include host name (id), total amount of CPU (Cs), amount of CPU usage (Cu), total amount of memory (Ms), amount of memory usage (Mu), total amount of hard disk (Hs), amount of hard disk usage (Hu), and the like. Since only the usage of the CPU, the memory, and the hard disk at a certain time cannot represent the usage within a certain period of time, for example, a certain time may be a peak value, but the rest of the time is unchanged. Therefore, the invention selects the time t1 and the time t2 to respectively obtain Cu1, Cu2, Mu1, Mu2, Hu1 and Hu2, and calculates the average utilization rate of the Cu1, the Cu2, the Mu1, the Mu2, the Hu1 and the Hu2 to represent the use conditions Cr and Mr of the related resources of the virtual machine in a period of timeAnd Hr. The time interval from t1 to t2 is generally fixed (i.e., the time interval from t1 to t2 is preset, and hereinafter collectively referred to as a preset time), and in this embodiment, the time interval is set to 30 seconds by default. If there is a special need, the administrator can adjust the time interval according to the need. Then, the average usage rates Cr ═ f (Cu1, Cu2)/Cs, Mr ═ f (Mu1, Mu2)/Ms, Hr ═ f (Hu1, Hu2)/Hs are calculated over the time t1 and t2, where f (x, y) ═ | x + y |/2. The average is not taken because the total amount of hard disks and the total amount of memory do not change during the program run. Constructing a feature vector xi from the featurest＝<id,Cs,Cr,Ms,Mr,Hs,Hr>. In the following description, Cr, Mr, and Hr are average usage rates in a predetermined time, and will not be described repeatedly.
2. In the embodiment, when the target virtual machine which does not meet the requirement is directly migrated, the related services in the virtual machine to be migrated are divided and partially migrated, so that the original virtual machine (i.e. the virtual machine to be migrated) can meet the operation requirement of one part of the services, and the other part of the services can normally operate in other target virtual machines which meet the requirement, and the migration is realized without affecting the normal operation of the services. In this embodiment, the service is divided into two parts (m is 2), and one part is migrated to one virtual machine for example, and may be migrated to a plurality of virtual machines if necessary, which is not described in detail here.
The comprehensive utilization rate Φ of the virtual machine is first calculated. The use conditions of the CPU, the memory and the hard disk are considered comprehensively, and the comprehensive use rate Φ is α · Cr + β · Mr + γ · Hr, where α, β, and γ are influence weights, that is, the attention of the administrator to the CPU, the memory and the hard disk, may be defined by the administrator according to different scenarios; in general, α, β, γ ∈ [0,1 ]]And α + β + γ ═ 1. Because the hard disk is generally easy to expand, the hard disk can be set with lower weight; and the memory and the CPU are used more, so the memory and the CPU are the core influencing the overall efficiency of the dynamic migration of the virtual machine, and higher weight can be set. In this embodiment, α, β, and γ are set to default values of α ═ 0.4, β ═ 0.4, and γ ═ 0.2, respectively. At the same time, the administrator defines a threshold Ψ, typically Ψ ∈ [0.5,1 ∈ [ ]]I.e. a virtualWhen the usage of the machine-related resources exceeds half, the virtual machine is in a heavy load state. When phi is>Ψ, a migration process needs to be initiated. Wherein
And is
The comprehensive utilization rate of each service in the virtual machine is obtained by calculation according to a phi calculation method.
Before migration starts, in the dividing process, the comprehensive utilization rate of each service in the virtual machine to be migrated needs to be counted, and the calculation is carried out according to a phi calculation method to obtain the result
Then sorting from large to small. In the classification process, the invention follows the principle of reducing the number of migration services as much as possible, so the virtual machines occupying the largest resources are accumulated, and when the occupied resources are more than half of the total number of the used resources and the number is the minimum, the virtual machines are divided into two classes, namely
And
the invention selects fewer services for migration, i.e. when i>When n/2, the migration Re becomes T2, and when i is not more than n/2, the migration Re becomes T1. Where Re is the partitioned set of services to be migrated. If the number of the migrated target virtual machines is greater than 2, the number of the classes of the service is divided into the corresponding number of the virtual machines, and details are not described here.
Judging whether a proper idle virtual machine is used as a target virtual machine or not, wherein the feature vector xi of the target virtual machinet＇＝<id＇,Cs＇,Cr＇,Ms＇,Mr＇,Hs＇,Hr＇>Need to satisfy { Cs' (1-Cr)>σ·Cs-Re·Cr-Re}∧{Ms＇·(1-Mr＇)>σ·Ms-Re·Mr-Re}∧{Hs＇·(1-Hr＇)>σ·Hs-Re·Hr-ReAnd (3) that the sizes of the idle CPU, the idle memory and the idle hard disk of the target virtual machine are more than sigma (sigma is more than or equal to 2) times of the utilization rate of a service set (namely the service set to be migrated) in the virtual machine to be migrated, so that the conditions are met. Wherein Cs-Re、Cr-Re、Ms-Re、Mr-Re、Hs-Re、Hr-ReRespectively representing the total amount and the utilization rate of a CPU, a memory and a hard disk in the service set Re to be migrated. If the appropriate virtual machine exists, the migration is carried out, and if the appropriate virtual machine does not exist, the migration is carried out after the service is classified again (the service is classified into m +1 classes).
3. In a specific migration process, it is necessary to comprehensively monitor related resources such as CPUs, memories, hard disks and the like related to the service set to be migrated Re ═ { s1, s2, …, sn }, and determine a migration sequence according to a change situation.
First, it is necessary to determine a change in the relevant resource of a service in a predetermined time, and the change in the relevant resource Θ si of the service si (i is 1, 2, … n) is determined<Cr-si,Mr-si,Hr-si>Wherein Cr is-si、Mr-si、Hr-siAnd (3) representing the average utilization rate of the CPU, the memory and the hard disk related to the service si, wherein the calculation method is the same as the step 1. If the Θ si is Φ, the CPU, the memory, and the hard disk are not changed, and the migration is performed directly. If the related resources of the service change, the services are sorted from small to large according to the comprehensive change rate of the related resources, and the sorted services Re' are migrated in sequence. Of course, the present invention gives different weights to each parameter, for example, when the influence of the memory change rate is considered to be larger than that of the hard disk, the memory is given a higher weight, and the corresponding values are calculated and compared. The invention adopts the influence weight of the comprehensive utilization rate phi of the virtual machine calculated in the step 2, and is convenient for comparison in a front-back unified way, namely the comprehensive change rate fc (Θ si) of the related resources of the service is alpha-Cc + beta-Mc + gamma-Hc.
In the migration process, the hard disk is set to be shared for storage, so that the interaction times and time are reduced conveniently in the data migration process; since the state of the CPU accounts for only a small portion of the migration, the migration process of the memory is mainly discussed. In the aspect of the memory, a pre-copy method is adopted, but the integral fast migration is carried out on the memory to be migrated by combining the shadow memory, so that the migration efficiency is improved. The shadow memory is mainly used for marking the memory change part by allocating a section of space to correspond to the corresponding part of the memory. The corresponding relationship between the shadow memory and the actual memory is mainly related to the memory change monitoring granularity, if the memory in migration is monitored and migrated according to the byte as a unit, 1 byte also corresponds to 1bit in the shadow memory, and the ratio is 8: 1. of course, in order to improve efficiency, the present invention monitors according to the size of the page, and 1 memory page (e.g. 4KB memory space) corresponds to 1bit in the shadow memory. When the monitored memory unit changes, the corresponding shadow memory corresponding part is marked as 1, otherwise, the shadow memory corresponding part is 0. In the subsequent migration, the copy migration of the portion marked as 1 in the shadow memory is performed. The concrete implementation is as follows:
firstly, the memory to be migrated is integrally copied to a target virtual machine. Then, corresponding shadow memory is divided in the original virtual machine (namely, the virtual machine to be migrated) to mark the memory change part from the beginning of migration. And dividing a part of idle memory in the memory part, uniformly pointing the changed memory part content to the idle memory, and then carrying out one-time migration on the divided memory part marked with change after the migration is finished once. And meanwhile, pointing the memory change part to the idle memory again, and circulating until the memory to be migrated is small enough and can be migrated within the time tau, suspending the virtual machine, and completing the whole migration process after one-time migration. Where τ may be defined by an administrator, typically a very small length of time, such as 0.01 s.
Referring to fig. 3, after the memory starts to migrate, the memory that has changed is marked by the shadow memory, such as 0x23F910, Ox23F914, Ox23FF2C, … …. After the whole migration is finished, the changed memories are mapped to a continuous area of the memory from a free area 0xFFFF00, and then the changed memory parts are migrated again. Meanwhile, the changed area in the original memory is marked again and mapped to a free area for migration. This process is cycled until the portion of memory that is changed is small enough to be migrated all at once within τ.
In summary, the present invention mainly solves the problem of providing a dynamic migration method for virtual machines based on service classification when the virtual machines have insufficient resources and there is no virtual machine with sufficient idle capacity for migration. The method monitors the comprehensive utilization rate of the virtual machine on the basis of acquiring the use condition of related resources of the virtual machine; and when the comprehensive utilization rate of the virtual machines is greater than a given threshold value, classifying the services in the virtual machines, selecting a proper virtual machine, and dynamically migrating related resources of part of the services to meet the use requirements of users. In addition, when there is a large enough idle virtual machine, the method can be directly used for dynamically migrating the virtual machine.
The above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and a person skilled in the art can make modifications or equivalent substitutions to the technical solution of the present invention without departing from the spirit and scope of the present invention, and the scope of the present invention should be determined by the claims.
Claims (4)
1. A dynamic migration method of virtual machines based on service classification comprises the following steps:
1) monitoring the use condition of related resources of each virtual machine in preset time through a virtual machine monitor, wherein the use condition of the related resources of the virtual machine comprises the sizes of a CPU (Central processing Unit), a memory and a hard disk and the average use ratio of the CPU, the memory and the hard disk in the preset time; acquiring the comprehensive utilization rate of the virtual machine; the comprehensive utilization rate phi is alpha, Cr, beta, Mr, gamma, Hr, wherein alpha, beta and gamma are influence weights and are defined by an administrator according to different scenes; cr, Mr and Hr respectively refer to the average utilization rate of a CPU, a memory and a hard disk within preset time;
2) when the comprehensive utilization rate of the virtual machine is greater than a threshold value, evaluating the comprehensive utilization rate of each service of the virtual machine, classifying each service according to the dividing requirement, and selecting a target virtual machine according to the classification result; wherein the threshold and partitioning requirements are defined by an administrator according to different scenarios; the size of the free space of the CPU, the memory and the hard disk of the target virtual machine is larger than sigma times of the utilization rate of the service set to be migrated, wherein sigma is larger than or equal to 2, and the service set to be migrated is divided according to the dividing requirement; when the comprehensive utilization rate of the virtual machine is greater than a threshold value, executing the following steps:
2-1) calculating the comprehensive utilization rate of each service in the virtual machine, sorting the services from large to small, and classifying the services according to a sorting result and a dividing requirement;
2-2) selecting a target virtual machine according to the classification result;
3) migrating the classified related resources of the partial services to a target virtual machine, wherein the steps comprise: migrating according to the sequence of the sequenced comprehensive utilization rate of each service, and determining the sequence of migration of related resources of the service according to the change condition of the related resources of the service in the monitored preset time; when the relevant resources of the service are not changed, directly carrying out migration; when the related resources of the service change, monitoring the changed service, sequencing according to the service change speed, and migrating according to the principle that the related resources of the service change from less to more; the related resources comprise host names, CPU total amount, CPU usage amount, memory total amount, memory usage amount, hard disk total amount and hard disk usage amount.
2. The method of claim 1, wherein step 2) further comprises: if no suitable target virtual machine exists, the services are reclassified.
3. The method of claim 1, wherein step 3) further comprises: the hard disk is set to shared storage.
4. The method of claim 1, wherein step 3) further comprises: intensively and rapidly migrating the memory by adopting a pre-copying method; wherein the concentrated fast migration of the pre-copy method refers to: marking the memory change part by using the shadow memory, copying the memory change part to an idle memory area, and continuously migrating until the content of the memory change part can be migrated at one time within tau time; where τ is defined by the administrator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710582510.8A CN107423114B (en) | 2017-07-17 | 2017-07-17 | Virtual machine dynamic migration method based on service classification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710582510.8A CN107423114B (en) | 2017-07-17 | 2017-07-17 | Virtual machine dynamic migration method based on service classification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107423114A CN107423114A (en) | 2017-12-01 |
| CN107423114B true CN107423114B (en) | 2020-12-22 |
Family
ID=60430105
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710582510.8A Active CN107423114B (en) | 2017-07-17 | 2017-07-17 | Virtual machine dynamic migration method based on service classification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107423114B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108549572A (en) * | 2018-04-08 | 2018-09-18 | 郑州云海信息技术有限公司 | A kind of heterogeneous server virtual management system and method for data center's total management system |
| CN108874535B (en) * | 2018-05-14 | 2022-06-10 | 中国平安人寿保险股份有限公司 | Task adjusting method, computer readable storage medium and terminal device |
| CN112948067A (en) * | 2019-12-11 | 2021-06-11 | 北京金山云网络技术有限公司 | Service scheduling method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335234A (en) * | 2015-10-29 | 2016-02-17 | 贵州电网有限责任公司电力调度控制中心 | Method for immediately migrating virtual machine |
| CN105607947A (en) * | 2015-12-11 | 2016-05-25 | 西北工业大学 | Novel cloud environment virtual machine scheduling method |
| CN106250210A (en) * | 2016-08-03 | 2016-12-21 | 上海交通大学 | Dispatching method of virtual machine under cloud environment |
| CN106406976A (en) * | 2016-07-21 | 2017-02-15 | 柏科数据技术（深圳）股份有限公司 | Method and apparatus for identifying IO intensive application in cloud computing environment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5644150B2 (en) * | 2010-03-23 | 2014-12-24 | 日本電気株式会社 | Service providing system, virtual machine server, service providing method, and service providing program |
-
2017
- 2017-07-17 CN CN201710582510.8A patent/CN107423114B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105335234A (en) * | 2015-10-29 | 2016-02-17 | 贵州电网有限责任公司电力调度控制中心 | Method for immediately migrating virtual machine |
| CN105607947A (en) * | 2015-12-11 | 2016-05-25 | 西北工业大学 | Novel cloud environment virtual machine scheduling method |
| CN106406976A (en) * | 2016-07-21 | 2017-02-15 | 柏科数据技术（深圳）股份有限公司 | Method and apparatus for identifying IO intensive application in cloud computing environment |
| CN106250210A (en) * | 2016-08-03 | 2016-12-21 | 上海交通大学 | Dispatching method of virtual machine under cloud environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107423114A (en) | 2017-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109923523B (en) | Computer system and method for computer system | |
| JP6219512B2 (en) | Virtual hadoop manager | |
| US9619378B2 (en) | Dynamically optimizing memory allocation across virtual machines | |
| EP2437168B1 (en) | Method and device for balancing load of multiprocessor system | |
| US9026630B2 (en) | Managing resources in a distributed system using dynamic clusters | |
| US9152200B2 (en) | Resource and power management using nested heterogeneous hypervisors | |
| US20140082202A1 (en) | Method and Apparatus for Integration of Virtual Cluster and Virtual Cluster System | |
| US9098337B2 (en) | Scheduling virtual central processing units of virtual machines among physical processing units | |
| WO2016078178A1 (en) | Virtual cpu scheduling method | |
| US9547510B2 (en) | Tracking guest memory characteristics for memory scheduling | |
| CN111078363A (en) | NUMA node scheduling method, device, equipment and medium for virtual machine | |
| US20160196157A1 (en) | Information processing system, management device, and method of controlling information processing system | |
| CN107423114B (en) | Virtual machine dynamic migration method based on service classification | |
| US20130097382A1 (en) | Multi-core processor system, computer product, and control method | |
| CN105389211A (en) | Memory allocation method and delay perception-memory allocation apparatus suitable for memory access delay balance among multiple nodes in NUMA construction | |
| US11797355B2 (en) | Resolving cluster computing task interference | |
| US20120054762A1 (en) | Scheduling apparatus and method for a multicore device | |
| US20190173770A1 (en) | Method and system for placement of virtual machines using a working set computation | |
| CN103106112A (en) | Method and device based on maximum load and used for load balancing scheduling | |
| US9588817B2 (en) | Scheduling method and scheduling system for assigning application to processor | |
| CN105930202B (en) | A kind of virtual machine migration method of three threshold values | |
| WO2024021475A1 (en) | Container scheduling method and apparatus | |
| Hwang et al. | Hyperdealer: Reference-pattern-aware instant memory balancing for consolidated virtual machines | |
| KR101639947B1 (en) | Hadoop preemptive deadline constraint scheduling method, execution program thereof method and recorded medium of the program | |
| Valencia et al. | Combining vm preemption schemes to improve vertical memory elasticity scheduling in clouds |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |