CN108920153A - A kind of Docker container dynamic dispatching method based on load estimation - Google Patents
A kind of Docker container dynamic dispatching method based on load estimation Download PDFInfo
- Publication number
- CN108920153A CN108920153A CN201810535697.0A CN201810535697A CN108920153A CN 108920153 A CN108920153 A CN 108920153A CN 201810535697 A CN201810535697 A CN 201810535697A CN 108920153 A CN108920153 A CN 108920153A
- Authority
- CN
- China
- Prior art keywords
- docker container
- load
- docker
- load estimation
- load data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/60—Software deployment
-
- 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
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
Abstract
The present invention provides a kind of Docker container dynamic dispatching method based on load estimation.This method is first by the load data of Docker container payload monitoring module acquisition Docker container;Load data is stored in time series databases again;Then Docker container payload prediction module obtains and handles load data from time series databases, in conjunction with ARIMA model analysis load data, and generates load estimation value;Last Docker container scheduler module extends the Docker container number in Docker container cluster according to load estimation value automatically.Load estimation technology is applied to Docker container scheduling field by the present invention, realize that Docker container cluster extends automatically according to real time load situation, effectively solve the problems, such as that Docker container cluster cannot improve the availability of Docker container cluster because loading condition adjust automatically resource is there are operation phase load capacity deficiency.
Description
Technical field
The invention belongs to Docker container dynamic dispatching technical field, especially it is to provide a kind of based on load estimation
Docker container dynamic dispatching method.
Background technique
One application is split into multiple independent, with service attribute services, each service operation by micro services framework
It in independent process, is cooperated with each other between service by the communication mechanism of lightweight, to provide business value for terminal user.
The deployment way of application based on micro services framework includes the modes such as manually dispose, script deployment based on cloud platform, and is based on
The mirror image deployment of Docker container is a kind of mode of current relatively mainstream.
As the application container engine of open source, Docker container enables developer application and its relies on encapsulation
To being then published to Docker container in transplantable Docker container, there are on the Linux machine of Docker container environment.
Docker container realizes the isolation of resource by the virtualization of operating system layer, and host sharing operation system, therefore can
The performance of greatly improve resource utilization and promote I/O etc..
In order to safeguard the availability of the application based on the deployment of Docker container, High Availabitity is constructed based on redundancy backup technology
Docker container cluster, and extending Docker container cluster by horizontal extension technology is common solution.Based on current
The Docker container cluster of existing Docker container layout deployment techniques building, can not be negative according to real-time Docker container
Situation adjust automatically resource is carried, causes Docker container cluster insufficient in operation phase load capacity.And dispose Docker container
Process include downloading mirror image, dispose mirror image, the starting sequence of operations such as Docker container.When the current Docker container money of discovery
Just start to apply for resource when source is not able to satisfy loading demand, since the operations such as downloading mirror image will lead to the new Docker container of deployment
Process expends the time very much, then, during this period, the availability of application will be unable to be guaranteed.
Therefore, the present invention provides a kind of Docker container dynamic dispatching method based on load estimation, and this Docker
Container dynamic dispatching method execution efficiency is high, reduces human intervention, is capable of the availability of efficiently maintenance application.
Summary of the invention
It is an object of the invention to load estimation technical application to Docker container dynamic dispatching, is solved current Docker
Container cluster cannot have operation phase load capacity deficiency because of loading condition adjust automatically resource, be Docker container
The user of cluster provides a kind of Docker container dynamic dispatching method based on load estimation.
The purpose of the present invention is realized at least through one of following technical solution.
A kind of Docker container dynamic dispatching method based on load estimation comprising:
The load data of Docker container is acquired first;Then load data is stored in time series databases;Then from when
Between sequence database obtain and processing load data, in conjunction with ARIMA model(Autoregressive Integrated Moving
Average model, ARMA model)Load data is analyzed, load estimation value is generated;Last basis is negative
Predicted value is carried, the Docker container number in Docker container cluster is extended.
Further, load data acquisition is completed by Docker container payload monitoring module, the monitoring of Docker container payload
Module according to the strategy of automatic regular polling acquired by way of inquiring pseudo-file Docker container to CPU, memory, magnetic disc i/o and
Load data of the consumption situation of network these fourth types resource as Docker container, then by load data and server node and
Docker information of container is associated, is stored in time series databases.
Further, the negative of Docker container is obtained from time series databases by Docker container payload prediction module
Data are carried, load data is abstracted in conjunction with ARIMA model, constructs online Docker container payload prediction model, then
It is analyzed by load data of the model to Docker container, generates load estimation value.
Further, the load for obtaining the generation of Docker container payload prediction module by Docker container scheduler module is pre-
Measured value is pressed when load estimation value is more than the load highest threshold value for the Docker container that Docker container cluster user specifies
According to the automatic extension container of step-length that Docker container cluster user specifies, until Docker container payload prediction module is given birth to again
At load estimation value be lower than Docker container load highest threshold value.
Further, by load estimation be applied to Docker container dynamic dispatching field realize Docker container cluster according to
Real time load situation extends automatically, and solving Docker container cluster cannot be because there are the operation phase for loading condition adjust automatically resource
The problem of load capacity deficiency improves the availability of Docker container cluster.
Compared with prior art, the invention has the advantages that and technical effect:
The present invention passes through load estimation technical application to Docker container dynamic dispatching field, Docker container cluster user
Simple configuration can realize that Docker container cluster extends automatically according to real time load situation, solve to be based on existing Docker
The Docker container cluster of container layout deployment techniques building cannot be negative there are the operation phase because of loading condition adjust automatically resource
The problem of loading capability deficiency improves the availability of Docker container cluster.
Detailed description of the invention
Fig. 1 is the Docker container dynamic dispatching method general frame figure based on load estimation;
Fig. 2 is Docker container payload prediction module work flow diagram;
Fig. 3 is Docker container scheduler module work flow diagram.
Specific embodiment
Specific implementation of the invention is described further below in conjunction with attached drawing and example, but implementation and protection of the invention
It is without being limited thereto, it is that those skilled in the art can refer to existing skill if place is not described in detail especially it is noted that having below
What art was realized.
This example provides a kind of Docker container dynamic dispatching method based on load estimation, general frame as shown in Figure 1,
Specifically include load data acquisition, load data storage, load data analysis and the dynamic dispatching of Docker container totally four processes.
Load data acquisition
Docker container cluster user disposes Docker container payload monitoring module in each server node first, and refers to
Determine the time interval of Docker container payload monitoring module acquisition load data, then starts Docker container payload and monitor mould
Block.
Docker container payload monitoring module acquires Docker container and provides in CPU, memory, magnetic disc i/o and network these fourth types
Load data of the Expenditure Levels in source as Docker container.In the cpu resource Expenditure Levels of Docker container, CPU index
It is divided into two classes:One kind embodies Docker container and consumes in the service condition of cpu resource, including user CPU consumption, system CPU, is each
A CPU consumption and CPU consumption in total;One kind embodies Docker container in the use degree of saturation of cpu resource, including CPU limitation
Execute number and CPU usage confined total time.What Docker container payload monitoring module acquired in terms of cpu resource consumption
It is service condition of the Docker container in cpu resource.
The core technology of Docker container is control group technology and NameSpace technology.Control group is by multiple control group subsystems
System composition realizes that wherein cpuacct subsystem is used to count the cpu usage of each control group, includes following three interface:
l cpuacct.stat:Report the CPU time that the control group is consumed in User space and kernel state respectively.
l cpuacct.usage:Report total CPU time of control group consumption.
l cpuacct.usage_percpu:Report the CPU time that the control group consumes on each CPU of host, always
That is the value of cpuacct.usage.
The interface of control group realizes that therefore, Docker container payload acquisition module passes through cpuacct based on pseudo file system
This control group subsystem, obtains real-time cpu load situation in a manner of file operation.
Docker container payload monitoring module is specified with the strategy of automatic regular polling according to Docker container cluster user
Time interval obtains the load data of Docker container by inquiry control group pseudo-file.
Load data storage
Collected load data is a new item to be recorded relevant with timestamp to Docker container payload monitoring module every time
Mesh needs to store these load datas, so that Docker container payload prediction module analyzes load data.Docker holds
Device load monitoring module stores load data using InfluxDB time series databases.The data of InfluxDB storage
Logically it is made of measurement, tag group, field group and a timestamp.
Measurement can be compared to be the tables of data in relevant database, Docker container payload monitoring module with
User CPU consumption, system CPU consumption, CPU consumption, memory source consumption, magnetic disc i/o resource consumption and Internet resources disappear in total
It consumes these load monitoring indexs and establishes measurement respectively, also represent the load data type of Docker container.
Tag group indicates that every records a series of corresponding attribute letters in database by one group of key-value pair data structure composition
Breath.Docker container cluster contains different server node and a large amount of Docker container, in order to obtain institute more quickly
For the load data of the Docker container needed to carry out data analysis and load estimation, Docker container payload monitoring module will
Server node name where Docker Container Name and Docker container is referred to as the tag group of load data, and Docker
Container Name serves not only as the unique identification of Docker container, has also contained the application service being deployed in the Docker container
Information.
Filed group is also to indicate the particular content of every record in database by key-value pair data structure composition.Docker
Specific load value of the container payload monitoring module in Field group storage Docker container.
Timestamp is used to record the generation time of the load data of Docker container, i.e. Docker container payload monitoring module
Collect the time of this load data.
Load data analysis
It is a time series that the timing that the load data of Docker container has, which represents the load data substantially,.To negative
The analysis and prediction for carrying data, are actually analyzed and predicted one group of time series data.ARIMA is currently used
Time series predicting model, in ARMA(Autoregressive moving average model)Upper introducing differential transformation,
Stationary time series is converted into original jiggly time series, data analysis is carried out with application arma modeling.Container payload
Data and time correlation connection, showing container payload data actually is also one group of time series data, and past container payload
Data can provide reference for the value range of the load estimation data of container future in a short time, so can be this by ARIMA
Time series predicting model carries out data analysis to container payload data and reasonable predicted value is calculated.
Docker container payload prediction module analyzes load data and the process of generation load estimation is as shown in Figure 2.
Docker container payload prediction module receives server node title, the Docker that Docker container cluster user specifies first
Container Name, needs predict load data type, analysis period and these master datas of the period of prediction, then from
InfluxDB time series databases obtain the load data of Docker container using these master datas as querying condition, raw
At time series.
Then, Docker container payload prediction module uses the method proof load data of unit root test to time series
Whether there is stationarity, if not having stationarity, difference operation is carried out to load data, until meeting stationarity item
Part.
Then the order p and order q of ARIMA model are determined using bayesian information criterion, and combine difference order, structure
Build the load estimation model of Docker container.Then, Docker container payload prediction module is used according to Docker container cluster
The predicted time section that person specifies, application load prediction model generate the load estimation value of Docker container during this period of time.
Finally, being needed if the load data of Docker container passes through differential transformation before constructing load estimation model
Restoring operation is carried out to load estimation value, can just obtain the load estimation value of final required Docker container.
Container dynamic dispatching
Docker container scheduler module realizes the stream of dynamic dispatching according to the load estimation value of Docker container to Docker container
Journey is as shown in Figure 3.Docker container scheduler module receives the Docker container that Docker container cluster user specifies first
Load highest threshold value and dilatation step-length.The load highest threshold value of Docker container is whether to carry out dynamic dispatching to Docker container
Criterion.When the load estimation value of Docker container is more than the load highest threshold value of Docker container, Docker container
Scheduler module assert that the load of the Docker container beyond limitation, can not support existing access request, it is necessary to via it
His Docker container shares the access request to service.Otherwise, Docker container scheduler module assert the Docker container still
The state good for use of service is maintained with sufficient machine resources.
Docker container scheduler module is 1 to 3 to dilatation step size settings, i.e. Docker container scheduler module exists every time
When Docker container cluster is that service increases Docker container copy, it is merely able to increase by 1 to 3 Docker container.This dilatation
The setting of step-length is that Docker container scheduler module disposably increases excessively in horizontal extension Docker container in order to prevent
Docker container causes a service to possess excessive Docker container copy, to cause the waste of machine resources.
When Docker container scheduler module, which determines, needs horizontal extension Docker container, select in Docker container cluster
Less server node is loaded, and the deployment request of Docker container is distributed to the node.It is opened in newly-increased Docker container
After moving successfully, Docker container scheduler module requests Docker container payload prediction module to the real-time negative of Docker container again
Data are carried to be analyzed and return to new load estimation value.Then, Docker container scheduler module compares load estimation value again
The load highest threshold value for the Docker container specified with Docker container cluster user judges whether to need to continue to extend
Docker container.
Claims (5)
1. a kind of Docker container dynamic dispatching method based on load estimation, it is characterised in that including:
The load data of Docker container is acquired first;Then load data is stored in time series databases;Then from when
Between sequence database obtain and processing load data, load data is analyzed in conjunction with ARIMA model, generate load estimation
Value;Finally according to load estimation value, the Docker container number in Docker container cluster is extended.
2. a kind of Docker container dynamic dispatching method based on load estimation according to claim 1, feature exist
It is completed in load data acquisition by Docker container payload monitoring module, Docker container payload monitoring module is according to timing wheel
The strategy of inquiry acquires Docker container to CPU, memory, magnetic disc i/o and network these fourth types resource by way of inquiring pseudo-file
Load data of the consumption situation as Docker container, then by load data and server node and Docker information of container
It is associated, it is stored in time series databases.
3. a kind of Docker container dynamic dispatching method based on load estimation according to claim 1, feature exist
In by Docker container payload prediction module from time series databases obtain Docker container load data, in conjunction with
ARIMA model is abstracted load data, constructs online Docker container payload prediction model, then passes through the model pair
The load data of Docker container is analyzed, and load estimation value is generated.
4. a kind of Docker container dynamic dispatching method based on load estimation according to claim 1, feature exist
In obtaining the load estimation value that Docker container payload prediction module generates by Docker container scheduler module, work as load estimation
When value is more than the load highest threshold value for the Docker container that Docker container cluster user specifies, according to Docker container cluster
The automatic extension container of the step-length that user specifies, until the load estimation value that Docker container payload prediction module regenerates is low
In the load highest threshold value of Docker container.
5. special according to a kind of described in any item Docker container dynamic dispatching methods of load estimation of claims 1 ~ 4
Sign is that load estimation is applied to Docker container dynamic dispatching field realizes Docker container cluster according to real time load feelings
Condition extends automatically, solve Docker container cluster cannot because loading condition adjust automatically resource there are operation phase load capacity not
The problem of foot improves the availability of Docker container cluster.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810535697.0A CN108920153B (en) | 2018-05-29 | 2018-05-29 | Docker container dynamic scheduling method based on load prediction |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810535697.0A CN108920153B (en) | 2018-05-29 | 2018-05-29 | Docker container dynamic scheduling method based on load prediction |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108920153A true CN108920153A (en) | 2018-11-30 |
CN108920153B CN108920153B (en) | 2022-04-22 |
Family
ID=64411128
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810535697.0A Active CN108920153B (en) | 2018-05-29 | 2018-05-29 | Docker container dynamic scheduling method based on load prediction |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108920153B (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111190790A (en) * | 2019-12-17 | 2020-05-22 | 西安交通大学 | Cloud computing cluster monitoring method and system based on peak prediction |
CN111327655A (en) * | 2018-12-14 | 2020-06-23 | 中移(杭州)信息技术有限公司 | Multi-tenant container resource quota prediction method and device and electronic equipment |
CN111385142A (en) * | 2020-02-14 | 2020-07-07 | 重庆邮电大学 | Kubernetes-based adaptive web container stretching method |
CN111427753A (en) * | 2020-03-23 | 2020-07-17 | 上海新炬网络信息技术股份有限公司 | ARIMA model-based capacity prediction device and control method thereof |
CN111625196A (en) * | 2020-05-26 | 2020-09-04 | 北京海益同展信息科技有限公司 | Block chain node capacity expansion method and device, computer equipment and storage medium |
CN111949374A (en) * | 2020-08-18 | 2020-11-17 | 广州市优普科技有限公司 | Intelligent processing platform of business data machine |
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 |
CN113672335A (en) * | 2021-07-08 | 2021-11-19 | 浙江一山智慧医疗研究有限公司 | Container scheduling method, device, electronic device and storage medium |
WO2022016808A1 (en) * | 2020-07-22 | 2022-01-27 | 中国科学院深圳先进技术研究院 | Kubernetes cluster resource dynamic adjustment method and electronic device |
WO2023051164A1 (en) * | 2021-09-30 | 2023-04-06 | International Business Machines Corporation | Resource reuse for pipeline workloads |
WO2024002190A1 (en) * | 2022-06-30 | 2024-01-04 | 中兴通讯股份有限公司 | Monitor-based container adjustment method and device, and storage medium |
US11977922B2 (en) | 2021-09-30 | 2024-05-07 | International Business Machines Corporation | Resource reuse for pipeline workloads |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103425535A (en) * | 2013-06-05 | 2013-12-04 | 浙江大学 | Agile elastic telescoping method in cloud environment |
CN106095533A (en) * | 2016-06-14 | 2016-11-09 | 中国联合网络通信集团有限公司 | Method of server expansion and device |
CN106790595A (en) * | 2016-12-29 | 2017-05-31 | 上海理想信息产业(集团)有限公司 | A kind of Docker containers active load balancer and method |
CN106961351A (en) * | 2017-03-03 | 2017-07-18 | 南京邮电大学 | Intelligent elastic telescopic method based on Docker container clusters |
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 |
CN107370816A (en) * | 2017-07-26 | 2017-11-21 | 郑州云海信息技术有限公司 | A kind of dispositions method and device of Web applications |
-
2018
- 2018-05-29 CN CN201810535697.0A patent/CN108920153B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103425535A (en) * | 2013-06-05 | 2013-12-04 | 浙江大学 | Agile elastic telescoping method in cloud environment |
CN106095533A (en) * | 2016-06-14 | 2016-11-09 | 中国联合网络通信集团有限公司 | Method of server expansion and device |
CN106790595A (en) * | 2016-12-29 | 2017-05-31 | 上海理想信息产业(集团)有限公司 | A kind of Docker containers active load balancer and method |
CN106961351A (en) * | 2017-03-03 | 2017-07-18 | 南京邮电大学 | Intelligent elastic telescopic method based on Docker container clusters |
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 |
CN107370816A (en) * | 2017-07-26 | 2017-11-21 | 郑州云海信息技术有限公司 | A kind of dispositions method and device of Web applications |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111327655A (en) * | 2018-12-14 | 2020-06-23 | 中移(杭州)信息技术有限公司 | Multi-tenant container resource quota prediction method and device and electronic equipment |
CN111190790A (en) * | 2019-12-17 | 2020-05-22 | 西安交通大学 | Cloud computing cluster monitoring method and system based on peak prediction |
CN111385142A (en) * | 2020-02-14 | 2020-07-07 | 重庆邮电大学 | Kubernetes-based adaptive web container stretching method |
CN111385142B (en) * | 2020-02-14 | 2022-04-12 | 重庆邮电大学 | Kubernetes-based adaptive web container stretching method |
CN111427753A (en) * | 2020-03-23 | 2020-07-17 | 上海新炬网络信息技术股份有限公司 | ARIMA model-based capacity prediction device and control method thereof |
CN111427753B (en) * | 2020-03-23 | 2024-04-23 | 上海新炬网络信息技术股份有限公司 | Capacity prediction device based on ARIMA model and control method thereof |
CN111625196A (en) * | 2020-05-26 | 2020-09-04 | 北京海益同展信息科技有限公司 | Block chain node capacity expansion method and device, computer equipment and storage medium |
WO2022016808A1 (en) * | 2020-07-22 | 2022-01-27 | 中国科学院深圳先进技术研究院 | Kubernetes cluster resource dynamic adjustment method and electronic device |
CN111949374B (en) * | 2020-08-18 | 2023-11-24 | 广州市优普科技有限公司 | Intelligent processing platform for business data machine |
CN111949374A (en) * | 2020-08-18 | 2020-11-17 | 广州市优普科技有限公司 | Intelligent processing platform of business data machine |
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 |
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 |
CN113672335A (en) * | 2021-07-08 | 2021-11-19 | 浙江一山智慧医疗研究有限公司 | Container scheduling method, device, electronic device and storage medium |
WO2023051164A1 (en) * | 2021-09-30 | 2023-04-06 | International Business Machines Corporation | Resource reuse for pipeline workloads |
US11977922B2 (en) | 2021-09-30 | 2024-05-07 | International Business Machines Corporation | Resource reuse for pipeline workloads |
WO2024002190A1 (en) * | 2022-06-30 | 2024-01-04 | 中兴通讯股份有限公司 | Monitor-based container adjustment method and device, and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN108920153B (en) | 2022-04-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108920153A (en) | A kind of Docker container dynamic dispatching method based on load estimation | |
CN105049268B (en) | Distributed computing resource distribution system and task processing method | |
CN104618693B (en) | A kind of monitor video based on cloud computing handles task management method and system online | |
CN110134495B (en) | Container cross-host online migration method, storage medium and terminal equipment | |
Rao et al. | Performance issues of heterogeneous hadoop clusters in cloud computing | |
CN106534318B (en) | A kind of OpenStack cloud platform resource dynamic scheduling system and method based on flow compatibility | |
CN102971724B (en) | The method and apparatus relevant with the management based on modular virtual resource in data center environment | |
CN109189589A (en) | A kind of distribution big data computing engines and framework method | |
CN104050042B (en) | The resource allocation methods and device of ETL operations | |
CN109918198A (en) | A kind of emulation cloud platform load dispatch system and method based on user characteristics prediction | |
CN106354729B (en) | Graph data processing method, device and system | |
CN104503832B (en) | A kind of scheduling virtual machine system and method for fair and efficiency balance | |
CN108182105A (en) | Local dynamic station moving method and control system based on Docker container techniques | |
EP3191948A1 (en) | Computing instance launch time | |
CN111381928B (en) | Virtual machine migration method, cloud computing management platform and storage medium | |
CN104243617A (en) | Task scheduling method and system facing mixed load in heterogeneous cluster | |
CN110661842A (en) | Resource scheduling management method, electronic equipment and storage medium | |
CN102929769A (en) | Virtual machine internal-data acquisition method based on agency service | |
Dong et al. | Energy-aware scheduling schemes for cloud data centers on google trace data | |
CN106101212A (en) | Big data access method under cloud platform | |
CN104580194A (en) | Virtual resource management method and device oriented to video applications | |
CN116450355A (en) | Multi-cluster model training method, device, equipment and medium | |
US11182189B2 (en) | Resource optimization for virtualization environments | |
CN116389591A (en) | Cross-domain-based distributed processing system and scheduling optimization method | |
CN116775420A (en) | Information creation cloud platform resource display and early warning method and system based on Flink flow calculation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |