CN107562545A - A kind of container dispatching method based on Docker technologies - Google Patents
A kind of container dispatching method based on Docker technologies Download PDFInfo
- Publication number
- CN107562545A CN107562545A CN201710809768.7A CN201710809768A CN107562545A CN 107562545 A CN107562545 A CN 107562545A CN 201710809768 A CN201710809768 A CN 201710809768A CN 107562545 A CN107562545 A CN 107562545A
- Authority
- CN
- China
- Prior art keywords
- resource
- weights
- container
- node
- formula
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Abstract
The invention discloses a kind of container dispatching method based on Docker technologies, increase information collection module in scheduler module, described information collection module is used to periodically obtain the resource allocation information of container in node, generates corresponding periodic time series；Calculate the static resource that container distributes in node and utilize weights;According to sequence for the previous period, Dynamic Weights forecast model is established, then dynamic resource is predicted using weights in node to latter cycle container；It is integrated ordered using weights progress with the dynamic resource using weights according to the static resource;According to the integrated ordered result, container deployment node is allocated；The new dynamic resource obtained according to periodicity time series utilizes weights, enters the resource allocation of Mobile state adjustment container.The present invention has the characteristics of can heightening resource utilization and keeping cluster resource equally loaded.
Description
Technical field
The present invention relates to a kind of container dispatching method based on Docker technologies.
Background technology
Docker is since increasing income, and just of great interest and discussion, Docker is an engine increased income, can
With easily for any application one lightweight of establishment, transplantable, self-centered container.Later stage has also issued ecosystem
Container cluster management instrument Swarm, for managing Docker clusters, make Docker clusters for a user when virtual in one
Entirety.The work that Swarm is mainly completed is：Container is operated on suitable node according to scheduling strategy, due on node
Run the difference of container, its resource utilization also difference.And the resource utilization of each node determines whole cluster
Loading condition.Therefore, the excellent summary of colony dispatching strategy is just particularly important.
The resource of different container demand different dimensions, when the resource exhaustion of any dimension of node, if more
The container of dimension resource requirement is activated, then the node will can not meet the needs of creating container, cannot also run this container.
In this case, the surplus resources of other dimensions just have been idle, and these unemployed resources are known as resource
Fragment, this is a kind of greatly waste.So just need to reduce the size of resource fragmentation.Meanwhile the equilibrium of cluster overall load
Situation determines general performance of the cluster in service, in order to improve the service quality of cluster, it is necessary to ensure the negative of whole cluster
Carry balanced.The content of the invention
Resource utilization can be heightened the technical problem to be solved by the invention is to provide one kind and keeps cluster resource equal
Weigh the container dispatching method loaded.
In order to solve the above technical problems, the technical solution adopted by the present invention is：
A kind of container dispatching method based on Docker technologies, increases information collection module in scheduler module, and described information is received
Collect module to be used to periodically obtain the resource allocation information of container in node, generate corresponding periodic time series；
Calculate the static resource that container distributes in node and utilize weights;
According to sequence for the previous period, Dynamic Weights forecast model is established, then latter cycle container is dynamically provided in node
Source is predicted using weights；
It is integrated ordered using weights progress with the dynamic resource using weights according to the static resource;
According to the integrated ordered result, container deployment node is allocated；
The new dynamic resource obtained according to periodicity time series utilizes weights, enters the resource allocation of Mobile state adjustment container.
The resource allocation information includes CPU, internal memory and the wide-band-message of container.
The static resource carries out assignment using weights using the distribution resource utilization variance.
The dynamic resource establishes Dynamic Weights forecast model using weights based on gray model, tries to achieve periodic dynamic
Weights.
The static resource is as follows using Weights-selected Algorithm：
Node resource dimension is D [1,2 ..., d], and the allocated resource of node is U [u1, u2... ud], node resource total amount is T
[t1, t2... td], node resource to be allocated is P [p1, p2... pd], node resource utilization rate is R [r1, r2... rd], according to formula
（1）Calculate the static resource utilization rate of each this container of node distribution：
（1）
According to formula（2）Calculate various dimensions average resource：
（2）
According to formula（3）Calculate each node resource configuration resource variance yields：
（3）
The Dynamic Weights forecast model method for building up is as follows:
Various dimensions resource utilization before acquisition node in the n moment, forms original time series
；
The accumulated method of formation of the original time series is generated into new sequence；
GM is established to the new sequence（1,1）The differential equation corresponding to model, sees below formula（4）,：
（4）
In formula, α is the grey number of development,Grey number is controlled for interior generation；
If, solved, obtained using least square method：
（5）
To B matrixes（6）With Y matrixes（7）Matrix operation is carried out, according to formula（5）Obtain the grey number α of development and interior generation controls grey number；
（6）
（7）
Grey number α will be developed and interior generation controls grey numberSubstitute into forecast model formula（8）In, according to forecast model formula（8）
Calculate the forecast model value of the distribution resource of time interval [1, n+1]；
（8）
Neighboring prediction model value is subtracted each other according to formula (9), draws predicted value value of the distribution resource in time interval [1, n+1]；
（9）
Residual test is carried out to forecast model according to formula (10), whether assessment models reach requirement；
（10）
Distribution resources value summation to next cycle draws Dynamic Weights.
The beneficial effect that the present invention is reached：The present invention combines static resource and utilizes weights with dynamic resource using weights,
Resource utilization can be significantly improved in node distribution, and in later stage dynamic adjustresources distribution, it is ensured that balanced it can collect
The dynamic load of group.
Embodiment
The invention will be further described below.Following examples are only used for the technical side for clearly illustrating the present invention
Case, and can not be limited the scope of the invention with this.
A kind of container dispatching method based on Docker technologies, increases information collection module, the letter in scheduler module
Cease collection module to be used to periodically obtain the resource allocation information of container in node, generate corresponding periodically time sequence
Row, the resource allocation information include CPU, internal memory and the wide-band-message of container,
Calculate the static resource that container distributes in node and utilize weights;The static resource uses the distribution using weights
Resource utilization variance carries out assignment, and static resource is as follows using Weights-selected Algorithm：
Node resource dimension is D [1,2 ..., d], and the allocated resource of node is U [u1, u2... ud], node resource total amount is T
[t1, t2... td], node resource to be allocated is P [p1, p2... pd], node resource utilization rate is R [r1, r2... rd], according to formula
（1）Calculate the static resource utilization rate of each this container of node distribution：
（1）
According to formula（2）Calculate various dimensions average resource：
（2）
According to formula（3）Calculate each node resource configuration resource variance yields：
（3）
According to sequence for the previous period, Dynamic Weights forecast model is established, then latter cycle container is dynamically provided in node
Source is predicted using weights；The dynamic resource establishes Dynamic Weights forecast model using weights based on gray model, tries to achieve
Periodic Dynamic Weights.
The Dynamic Weights forecast model method for building up is as follows:
Various dimensions resource utilization before acquisition node in the n moment, forms original time series
；
The accumulated method of formation of the original time series is generated into new sequence；
GM is established to the new sequence（1,1）The differential equation corresponding to model, sees below formula（4）,：
（4）
In formula, α is the grey number of development,Grey number is controlled for interior generation；
If, solved, obtained using least square method：
（5）
To B matrixes（6）With Y matrixes（7）Matrix operation is carried out, according to formula（5）Obtain the grey number α of development and interior generation controls grey number；
（6）
（7）
Grey number α will be developed and interior generation controls grey numberSubstitute into forecast model formula（8）In, according to forecast model formula（8）
Calculate the forecast model value of the distribution resource of time interval [1, n+1]；
（8）
Neighboring prediction model value is subtracted each other according to formula (9), draws predicted value value of the distribution resource in time interval [1, n+1]；
（9）
Residual test is carried out to forecast model according to formula (10), whether assessment models reach requirement；
（10）
Distribution resources value summation to next cycle draws Dynamic Weights.
It is integrated ordered using weights progress with the dynamic resource using weights according to the static resource;It can use and divide
Minor sort either carries out weight and adds and sort, and according to the integrated ordered result, container deployment node is allocated；
The new dynamic resource obtained according to periodicity time series utilizes weights, enters the resource allocation of Mobile state adjustment container.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, without departing from the technical principles of the invention, some improvement and deformation can also be made, these are improved and deformation
Also it should be regarded as protection scope of the present invention.
Claims (6)
1. a kind of container dispatching method based on Docker technologies, it is characterized in that, increase information mould in scheduler module
Block, described information collection module are used to periodically obtain the resource allocation information of container in node, generate the corresponding cycle
The time series of property；
Calculate the static resource that container distributes in node and utilize weights;
According to sequence for the previous period, Dynamic Weights forecast model is established, then latter cycle container is dynamically provided in node
Source is predicted using weights；
It is integrated ordered using weights progress with the dynamic resource using weights according to the static resource;
According to the integrated ordered result, container deployment node is allocated；
The new dynamic resource obtained according to periodicity time series utilizes weights, enters the resource allocation of Mobile state adjustment container.
2. a kind of container dispatching method based on Docker technologies according to claim 1, it is characterized in that, the resource
Distribution information includes CPU, internal memory and the wide-band-message of container.
3. a kind of container dispatching method based on Docker technologies according to claim 1, it is characterized in that, the static state
Utilization of resources weights carry out assignment using the distribution resource utilization variance.
4. a kind of container dispatching method based on Docker technologies according to claim 1, it is characterized in that, the dynamic
Utilization of resources weights establish Dynamic Weights forecast model based on gray model, try to achieve periodic Dynamic Weights.
5. a kind of container dispatching method based on Docker technologies according to claim 3, it is characterized in that, the static state
Utilization of resources Weights-selected Algorithm is as follows：
Node resource dimension is D [1,2 ..., d], and the allocated resource of node is U [u1, u2... ud], node resource total amount is T
[t1, t2... td], node resource to be allocated is P [p1, p2... pd], node resource utilization rate is R [r1, r2... rd], according to formula
（1）Calculate the static resource utilization rate of each this container of node distribution：
（1）
According to formula（2）Calculate various dimensions average resource：
（2）
According to formula（3）Calculate each node resource configuration resource variance yields：
（3）.
6. a kind of container dispatching method based on Docker technologies according to claim 4, it is characterized in that, the dynamic
Weights forecast model method for building up is as follows：
Various dimensions resource utilization before acquisition node in the n moment, forms original time series
；
The accumulated method of formation of the original time series is generated into new sequence；
GM is established to the new sequence（1,1）The differential equation corresponding to model, sees below formula（4）,
（4）
In formula, α controls grey number to develop grey number, for interior generation；
If, solved, obtained using least square method：
（5）
To B matrixes（6）With Y matrixes（7）Matrix operation is carried out, according to formula（5）Obtain the grey number α of development and interior generation control ash
Number；
（6）
（7）
Grey number α will be developed and interior generation controls grey number to substitute into forecast model formula（8）In, according to forecast model formula（8）Meter
Calculate the forecast model value of the distribution resource of time interval [1, n+1]；
（8）
Neighboring prediction model value is subtracted each other according to formula (9), draws predicted value value of the distribution resource in time interval [1, n+1]；
（9）
Residual test is carried out to forecast model according to formula (10), whether assessment models reach requirement；
（10）
Distribution resources value summation to next cycle draws Dynamic Weights.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710809768.7A CN107562545A (en) | 2017-09-11 | 2017-09-11 | A kind of container dispatching method based on Docker technologies |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710809768.7A CN107562545A (en) | 2017-09-11 | 2017-09-11 | A kind of container dispatching method based on Docker technologies |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107562545A true CN107562545A (en) | 2018-01-09 |
Family
ID=60980541
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710809768.7A Pending CN107562545A (en) | 2017-09-11 | 2017-09-11 | A kind of container dispatching method based on Docker technologies |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107562545A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108415772A (en) * | 2018-02-12 | 2018-08-17 | 腾讯科技（深圳）有限公司 | A kind of resource adjusting method, device and medium based on container |
| CN108933834A (en) * | 2018-07-18 | 2018-12-04 | 郑州云海信息技术有限公司 | A kind of dispatching method and dispatching device |
| CN109117265A (en) * | 2018-07-12 | 2019-01-01 | 北京百度网讯科技有限公司 | The method, apparatus, equipment and storage medium of schedule job in the cluster |
| CN109144727A (en) * | 2018-08-21 | 2019-01-04 | 郑州云海信息技术有限公司 | The management method and device of resource in cloud data system |
| CN109582461A (en) * | 2018-11-14 | 2019-04-05 | 中国科学院计算技术研究所 | A kind of calculation resource disposition method and system for linux container |
| CN109656713A (en) * | 2018-11-30 | 2019-04-19 | 河海大学 | A kind of container dispatching method based on edge calculations frame |
| CN109981396A (en) * | 2019-01-22 | 2019-07-05 | 平安普惠企业管理有限公司 | The monitoring method and device, medium and electronic equipment of docker service container cluster |
| CN110990160A (en) * | 2019-12-27 | 2020-04-10 | 广西电网有限责任公司 | Static security analysis container cloud elastic expansion method based on load prediction |
| CN111158908A (en) * | 2019-12-27 | 2020-05-15 | 重庆紫光华山智安科技有限公司 | Kubernetes-based scheduling method and device for improving GPU utilization rate |
| CN111273871A (en) * | 2020-01-19 | 2020-06-12 | 星辰天合（北京）数据科技有限公司 | Method and device for dynamically allocating storage resources on container platform |
| CN111367632A (en) * | 2020-02-14 | 2020-07-03 | 重庆邮电大学 | Container cloud scheduling method based on periodic characteristics |
| CN112187894A (en) * | 2020-09-17 | 2021-01-05 | 杭州谐云科技有限公司 | Container dynamic scheduling method based on load correlation prediction |
| CN113051067A (en) * | 2019-12-27 | 2021-06-29 | 顺丰科技有限公司 | Resource allocation method, device, computer equipment and storage medium |
| WO2023066035A1 (en) * | 2021-10-18 | 2023-04-27 | 阿里巴巴（中国）有限公司 | Resource allocation method and resource allocation apparatus |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105488307A (en) * | 2016-01-13 | 2016-04-13 | 武汉福天通科技有限公司 | Evaluation method of slope monitoring and early warning system based on Big Dipper |
| CN106453492A (en) * | 2016-08-30 | 2017-02-22 | 浙江大学 | Docker container cloud platform container scheduling method based on fuzzy mode recognition |
| CN107045455A (en) * | 2017-06-19 | 2017-08-15 | 华中科技大学 | A kind of Docker Swarm cluster resource method for optimizing scheduling based on load estimation |
-
2017
- 2017-09-11 CN CN201710809768.7A patent/CN107562545A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105488307A (en) * | 2016-01-13 | 2016-04-13 | 武汉福天通科技有限公司 | Evaluation method of slope monitoring and early warning system based on Big Dipper |
| CN106453492A (en) * | 2016-08-30 | 2017-02-22 | 浙江大学 | Docker container cloud platform container scheduling method based on fuzzy mode recognition |
| CN107045455A (en) * | 2017-06-19 | 2017-08-15 | 华中科技大学 | A kind of Docker Swarm cluster resource method for optimizing scheduling based on load estimation |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108415772A (en) * | 2018-02-12 | 2018-08-17 | 腾讯科技（深圳）有限公司 | A kind of resource adjusting method, device and medium based on container |
| CN108415772B (en) * | 2018-02-12 | 2022-02-18 | 腾讯科技（深圳）有限公司 | Container-based resource adjustment method, device and medium |
| CN109117265A (en) * | 2018-07-12 | 2019-01-01 | 北京百度网讯科技有限公司 | The method, apparatus, equipment and storage medium of schedule job in the cluster |
| CN108933834A (en) * | 2018-07-18 | 2018-12-04 | 郑州云海信息技术有限公司 | A kind of dispatching method and dispatching device |
| CN109144727A (en) * | 2018-08-21 | 2019-01-04 | 郑州云海信息技术有限公司 | The management method and device of resource in cloud data system |
| CN109582461A (en) * | 2018-11-14 | 2019-04-05 | 中国科学院计算技术研究所 | A kind of calculation resource disposition method and system for linux container |
| CN109656713A (en) * | 2018-11-30 | 2019-04-19 | 河海大学 | A kind of container dispatching method based on edge calculations frame |
| CN109656713B (en) * | 2018-11-30 | 2022-09-16 | 河海大学 | Container scheduling method based on edge computing framework |
| CN109981396A (en) * | 2019-01-22 | 2019-07-05 | 平安普惠企业管理有限公司 | The monitoring method and device, medium and electronic equipment of docker service container cluster |
| CN109981396B (en) * | 2019-01-22 | 2022-07-08 | 平安普惠企业管理有限公司 | Monitoring method and device for cluster of docker service containers, medium and electronic equipment |
| CN111158908A (en) * | 2019-12-27 | 2020-05-15 | 重庆紫光华山智安科技有限公司 | Kubernetes-based scheduling method and device for improving GPU utilization rate |
| CN110990160A (en) * | 2019-12-27 | 2020-04-10 | 广西电网有限责任公司 | Static security analysis container cloud elastic expansion method based on load prediction |
| CN111158908B (en) * | 2019-12-27 | 2021-05-25 | 重庆紫光华山智安科技有限公司 | Kubernetes-based scheduling method and device for improving GPU utilization rate |
| CN113051067A (en) * | 2019-12-27 | 2021-06-29 | 顺丰科技有限公司 | Resource allocation method, device, computer equipment and storage medium |
| CN111273871A (en) * | 2020-01-19 | 2020-06-12 | 星辰天合（北京）数据科技有限公司 | Method and device for dynamically allocating storage resources on container platform |
| CN111367632A (en) * | 2020-02-14 | 2020-07-03 | 重庆邮电大学 | Container cloud scheduling method based on periodic characteristics |
| CN111367632B (en) * | 2020-02-14 | 2023-04-18 | 重庆邮电大学 | Container cloud scheduling method based on periodic characteristics |
| CN112187894B (en) * | 2020-09-17 | 2022-06-10 | 杭州谐云科技有限公司 | Container dynamic scheduling method based on load correlation prediction |
| CN112187894A (en) * | 2020-09-17 | 2021-01-05 | 杭州谐云科技有限公司 | Container dynamic scheduling method based on load correlation prediction |
| WO2023066035A1 (en) * | 2021-10-18 | 2023-04-27 | 阿里巴巴（中国）有限公司 | Resource allocation method and resource allocation apparatus |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107562545A (en) | A kind of container dispatching method based on Docker technologies | |
| CN102624922B (en) | Method for balancing load of network GIS heterogeneous cluster server | |
| CN105491138B (en) | Distributed load scheduling method based on load rate graded triggering | |
| CN108628674A (en) | Method for scheduling task, cloud platform based on cloud platform and computer storage media | |
| Liu et al. | Task scheduling in fog enabled Internet of Things for smart cities | |
| CN106385468A (en) | Method for balancing predictable dynamic load of Web clusters | |
| CN103401939A (en) | Load balancing method adopting mixing scheduling strategy | |
| CN103297499A (en) | Scheduling method and system based on cloud platform | |
| CN102664814A (en) | Grey-prediction-based adaptive dynamic resource allocation method for virtual network | |
| CN102035737A (en) | Adaptive load balancing method and device based on cognitive network | |
| CN104994156A (en) | Load balancing method and system for cluster | |
| CN104202254A (en) | An intelligent load balancing method based on a cloud calculation platform server | |
| CN107426332A (en) | The load-balancing method and system of a kind of web server cluster | |
| CN111124531A (en) | Dynamic unloading method for calculation tasks based on energy consumption and delay balance in vehicle fog calculation | |
| CN107193640B (en) | Virtual machine initial placement strategy method based on Rendezvous hash algorithm | |
| CN102541655A (en) | Method and device for implementing dynamic load balancing | |
| CN106998340B (en) | Load balancing method and device for board resources | |
| CN107832129A (en) | A kind of dynamic task scheduling optimization method of Based on Distributed stream calculation system | |
| CN106708625A (en) | Minimum-cost maximum-flow based large-scale resource scheduling system and minimum-cost maximum-flow based large-scale resource scheduling method | |
| CN104283963B (en) | A kind of CDN load-balancing methods of Distributed Cooperative formula | |
| CN104881322A (en) | Method and device for dispatching cluster resource based on packing model | |
| CN108427602A (en) | A kind of coordinated dispatching method and device of distributed computing task | |
| CN109284336B (en) | Geographically distributed data center system and scheduling method thereof | |
| CN113157431B (en) | Computing task copy distribution method for edge network application environment | |
| CN105187488A (en) | Method for realizing MAS (Multi Agent System) load balancing based on genetic algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200228 Address after: Room 2101, 2102 and 2103, 21 / F, Youzhi building, No.29 Kejian Road, Jiangning District, Nanjing City, Jiangsu Province Applicant after: Nanjing shouchang Information Engineering Co., Ltd Address before: 210000 Jiangsu Province, Nanjing District of Jiangning City, Mo Zhou Road No. 12 Building No. 2 r536 ugood Applicant before: Nanjing Austrian cloud Information Technology Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180109 |
|
| RJ01 | Rejection of invention patent application after publication |