CN110489200B - Task scheduling method suitable for embedded container cluster - Google Patents
Task scheduling method suitable for embedded container cluster Download PDFInfo
- Publication number
- CN110489200B CN110489200B CN201810457653.0A CN201810457653A CN110489200B CN 110489200 B CN110489200 B CN 110489200B CN 201810457653 A CN201810457653 A CN 201810457653A CN 110489200 B CN110489200 B CN 110489200B
- Authority
- CN
- China
- Prior art keywords
- node
- resource
- nodes
- parameter
- calculating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/4557—Distribution of virtual machine instances; Migration and load balancing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
- G06F2009/45583—Memory management, e.g. access or allocation
Abstract
The invention discloses a task scheduling method suitable for an embedded container cluster, which comprises the following steps: calculating a resource use balance parameter, a load balance parameter and an operation stability parameter of each node of the container cluster; the resource usage balance parameter is the resource usage rate variance of each dimensionality of the node; the load balancing parameter is a resource load weight obtained by linearly weighting the resource utilization rate of each dimension of the use node; the stability parameter is a node stability value obtained according to the node online rate and the task completion rate; and respectively generating three sorted lists according to the resource use balance parameter, the load balance parameter and the stability parameter of each node, taking the position values of the nodes in the three sorted lists as three-dimensional space coordinates, and selecting the node with the minimum Chebyshev distance to the original point to deploy a new container. The method of the invention can reduce the influence of the dynamic change of the node on the container deployment strategy and improve the overall performance of the container cluster.
Description
Technical Field
The invention relates to the field of computers, in particular to the field of container cluster scheduling, and particularly relates to a task scheduling method suitable for an embedded container cluster.
Background
With the rapid development of the internet and the internet of things, information networks in the future form a ternary fusion information world integrating human, machine and thing, massive heterogeneous terminals and large-scale data processing, and great challenges are brought to the processing capacity of a traditional service system. The currently proposed technical architecture of the sea service can realize the near field resource aggregation and provision of the user side sea end node. Therefore, the method has great significance for realizing effective management and regulation of the sea-end embedded resources.
The docker is a lightweight container virtualization technology, and can conveniently realize the deployment and migration of tasks in a heterogeneous cluster. The individual docker software cannot meet the scheduling requirements of the cluster. To solve the problem, no matter IT is a huge head, a startup company, or a common enterprise user, an integrated docker container, i.e., a service management platform, is needed, so that the user can transparently enjoy the convenience brought by the docker container, and finally, the purpose of enabling the applications of the user to be smoothly clouded is achieved. The development of containers as a service is also very diverse as an important component of the docker ecosphere, including swarm, kubernets, messs, aws, ecs, and the like.
At present, the scheduling method of embedded container cluster management software has a plurality of problems, firstly, each dimension resource of a container cannot be well utilized, the maximum potential of a cluster cannot be exerted, and the user reaction is not very good; secondly, the utilization rate of system resources is low, and the node dynamic performance is high; the main reason for the above problem is that the node resource fragmentation and the cluster load are not balanced.
Disclosure of Invention
The invention aims to solve the problems of unbalanced resource load, low system resource utilization rate, high node dynamic property and the like in an embedded container cluster, and provides a task scheduling method suitable for the embedded container cluster.
In order to achieve the above object, the present invention provides a task scheduling method suitable for an embedded container cluster, where the method includes:
calculating a resource use balance parameter, a load balance parameter and an operation stability parameter of each node of the container cluster; the resource usage balance parameter is the resource usage rate variance of each dimensionality of the node; the load balancing parameter is a resource load weight obtained by linearly weighting the resource utilization rate of each dimension of the use node; the stability parameter is a node stability value obtained according to the node online rate and the task completion rate;
and respectively generating three sorted lists according to the resource use balance parameter, the load balance parameter and the stability parameter of each node, taking the position values of the nodes in the three sorted lists as three-dimensional space coordinates, and selecting the node with the minimum Chebyshev distance to the original point to deploy a new container.
As an improvement of the above method, the method specifically comprises:
step 1) obtaining node diI is more than or equal to 1 and less than or equal to K, and calculating the resource utilization rate variance of each node; the nodes are arranged in an ascending order according to the variance of the resource utilization rate to obtain a sequence L1；
Step 2) calculating the resource load weight of each node according to the resource utilization rate of each dimension obtained in the step 1); according to the resource load weight, the nodes are arranged in an ascending order to obtain a sequence L2；
Step 3) calculating a stable value of each node according to the historical record of each node; according to the stable value, the nodes are arranged in ascending order to obtain a sequence L3；
Step 4) three sequences L according to nodes1、L2And L3The position value in (3) maps the node to a three-dimensional space, and the Chebyshev distance dis value of the node is calculated:
and 5) selecting the node with the minimum dis value as an optimal node, and deploying the task to the node.
As an improvement of the above method, the step 1) specifically includes:
step 1-1) computing node diMemory usage rate M (d)i)：
M(di)＝((MenTotal-MemFree)/MemTotal)×100％
Wherein, MemTotal is the total memory amount, and MemFree is the memory margin;
step 1-2) computing node diCpu usage of C (d)i)：
C(di)＝((us+sy)/(us+sy+id))×100％
Wherein us is a user state, sy is a kernel state, and id is an idle state;
step 1-3) calculating node diNetwork load usage rate of N (d)i)：
N(di)＝((N1+N2)/M)×100％
Wherein, N1 is the incoming flow, N2 is the outgoing flow, and M is the maximum network throughput;
step 1-4) calculating node diAverage resource usage of (2):
wherein N is 3; r is1i＝M(di)，r2i＝C(di)，r3i＝N(di)；
Step 1-5) calculating the node d according to the average resource utilization rateiResource usage variance of (2):
step 1-6) according to the variance S of the resource utilization ratei 2Carrying out ascending arrangement on the K nodes to obtain a sequence L1。
As an improvement of the above method, the step 2) specifically includes:
step 2-1) obtaining the memory utilization rate M (d)i) Cpu utilization C (d)i) And network load usage rate N (d)i) Weight k ofj，1≤j≤N；
Step 2-2 computing node diResource load weight W (d)i)：
Wherein, γjTo adjust the coefficients of the weights, R1(di)＝M(di)，R2(di)＝C(di)，R3(di)＝N(di) (ii) a p is a threshold;
step 2-3) according to the resource load weight W (d)i) Carrying out ascending arrangement on the K nodes to obtain a sequence L2。
As an improvement of the above method, the step 1) specifically includes:
step 3-1) a computing node diA stable value of Z (d)i)：
Wherein M is the number of unit time slices; n isjNumber of tasks completed in a unit time slice for a node, NjReceiving the total number of tasks, t, for a unit time slice of a nodejIs the on-line time length in the unit time slice of the node, tpIs a unit time slice;
step 3-2) based on the stabilized value Z (d)i) Carrying out ascending arrangement on the K nodes to obtain a sequence L3。
As an improvement of the above method, the step 4) is specifically:
calculating the Chebyshev distance dis from the node to the node in the three-dimensional space according to the following formula:
wherein, ci1,ci2,ci3Are respectively node diIn the sequence L1，L2，L3The position value of (1).
The invention has the advantages that:
1. the method can reduce the influence of the dynamic change of the nodes on the container deployment strategy and improve the overall performance of the container cluster;
2. the method of the invention provides dynamic weight to improve the resource utilization rate of the container cluster, fully utilizes all dimension resources of the nodes, and simultaneously, the containers are deployed in a balanced manner to the maximum extent, thereby improving the usability of the cluster and exerting the potential of the cluster;
3. the method divides the nodes into a high stability set and a low stability set, and preferentially deploys the tasks on the nodes in the high stability set, so that the method can be better applied to the embedded cluster with stronger activity.
Drawings
Fig. 1 is a task scheduling system architecture diagram according to embodiment 1 of the present invention;
fig. 2 is a flowchart of a container cluster task scheduling method according to embodiment 2 of the present invention.
Detailed Description
In order to make the technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the present invention are further described in detail below with reference to the accompanying drawings and the embodiments.
The technical scheme of the invention comprehensively considers the problems of high node dynamic property, load balance and resource fragmentation of the cluster, and simultaneously optimizes the method aiming at the characteristic of high dynamic property of the embedded cluster. And aiming at the resource fragments generated in the cluster operation process, the resource variance is used for measuring the equilibrium condition of the resource use on the single node. According to the variance of the resource utilization rate, the nodes are arranged in an ascending order to obtain a sequence L1(ii) a Meanwhile, aiming at the problem of cluster load balance, a dynamic weight algorithm is adopted, and multidimensional parameter description node weights such as a memory, a cpu and a network load are selected. And giving corresponding weight to each parameter, and calculating the weight of each node, wherein the smaller the weight is, the lighter the load of the node is. According to the weight value, the nodes are arranged in an ascending order to obtain a sequence L2. In order to more accurately reflect the node load condition, when the utilization rate of a certain resource dimension exceeds a set threshold value, the weight of the dimension resource is actively increased. For the problem of high dynamic property, the invention provides a method for calculating a node stable value. The calculation method is based on the node online rate and the task completion rate. According to the size of the stable value, the nodes are arranged in an ascending order to obtain L3And taking the position values of the nodes in the three sequences as three-dimensional space coordinates, and mapping the nodes into a three-dimensional space. The container is deployed at the node with the smallest chebyshev distance to the origin.
Example 1
As shown in fig. 1, embodiment 1 of the present invention provides a task scheduling system including: the system comprises a sweep cluster module, a weight module and a scheduling module.
The system comprises a sweep cluster module, a service discovery module and a service discovery module, wherein the sweep cluster module is used for providing a basic operation environment of a cluster system and is used for realizing interaction with a client, interaction with a working node, cluster scheduling and service discovery; 5 worker nodes are deployed in the cluster.
The number of working nodes is K-5, and the number of nodes is diI is more than or equal to 1 and less than or equal to K, the resource dimension is N, and the resource utilization rate of each node is (r)1,…rN)。
The weight module is used for regularly acquiring the resource use information of each node, calculating the weight of each node in the cluster according to the current resource use condition of the node and the resource required by the new container, and feeding back the optimal node to the scheduling module;
and the scheduling module is used for deploying the new container on the optimal node according to the feedback information of the weight module.
As shown in fig. 2, embodiment 2 of the present invention provides a task scheduling method suitable for an embedded container cluster, where the method includes:
And providing information required by container starting and a container mirror image to the swarm cluster management program through the docker client.
Specifically, the memory usage rate M (d)i) Obtaining the residual quantity MemFree and the total quantity of memory MemTotal under a linux system/proc/meminfo directory.
Memory usage M (d)i) The expression is as follows:
M(di)＝((MenTotal-MemFree)/MemTotal)×100％
cpu utilization C (d)i) And returning the user state us, the kernel state sy and the idle state id in the value by adopting the top command under linux. The cpu usage expression is:
C(di)＝((us+sy)/(us+sy+id))×100％
if the inflow traffic is N1, the outflow traffic is N2, and the maximum network throughput is M, the network load usage rate N (d)i) The expression is as follows:
N(di)＝((N1+N2)/M)×100％
Computing node diAverage resource usage of (2):
wherein N is 3; r is1i＝M(di)，r2i＝C(di)，r3i＝N(di)；
Step 1-5) calculating the node d according to the average resource utilization rateiResource usage variance Si 2：
for the problem of load balancing, the embodiment proposes that the ratio of the memory resources which have been allocated for the container but are not used by the node to the total memory of the node is used as a new resource dimension, so as to improve the accuracy of the load measurement of the memory resources of the node. The use of dynamic weights is also proposed. Due to the fact that each dimension resource is weighted, the situation that the occupation ratio of a certain dimension is high and the node weight cannot be reflected is caused. Therefore, in order to further improve the evaluation capability of the algorithm, the weight of the corresponding parameter is dynamically adjusted.
Firstly, setting a threshold value for each dimension, and increasing the resource weight to be k times of the original weight when the resource utilization rate of the dimension reaches the threshold value, wherein k can be self-owned by a userIs already specified. Preferred setting γjIs 2. Although the resource load of the node is improved to a certain extent, the actual situation is reflected more reasonably. Calculating the resource load weight W (d) of the node according to the following formulai)：
Wherein k isjWeight values, gamma, representing resources of respective dimensionsjCoefficient for readjusting weight according to upper limit of load, R1(di)＝M(di)，R2(di)＝C(di)，R3(di)＝N(di) (ii) a p is a threshold;
The stability rating division rule is as follows:
wherein M is the number of unit time slices; n isjNumber of tasks completed in a unit time slice for a node, NjReceiving the total number of tasks, t, for a unit time slice of a nodejIs the on-line time length in the unit time slice of the node, tpIs a unit time slice; z (d)i) Is a node stable value; when in use
A high-stability node is present, otherwise a low-stability node,
is a stability threshold.
calculating the Chebyshev distance dis from the node to the node in the three-dimensional space according to the following formula:
wherein, ci1,ci2,ci3Are respectively diAt L1，L2，L3The position value of (1).
And step 210, judging whether a multi-node dis value is minimum in parallel, if not, jumping to step 211, and if so, jumping to step 212.
And step 211, selecting the node deployment container with the minimum dis value.
And step 212, randomly selecting a node deployment container with the minimum parallel dis value.
The method provided by the invention comprehensively considers two typical problems of resource fragmentation and load balancing in the cluster. The dynamic weight is provided, the utilization rate of container cluster resources is improved, the container is deployed in a balanced manner to the maximum extent while all dimension resources of the nodes are fully utilized, the usability of the cluster is improved, and the potential of the cluster is exerted. And dividing the nodes into a high stability set and a low stability set, and preferentially deploying the tasks on the nodes in the high stability set, so that the method can be better applied to the embedded cluster with strong activity.
Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention and are not limited. Although the present invention has been described in detail with reference to the embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (5)
1. A task scheduling method for an embedded container cluster, the method comprising:
calculating a resource use balance parameter, a load balance parameter and an operation stability parameter of each node of the container cluster; the resource usage balance parameter is the resource usage rate variance of each dimensionality of the node; the load balancing parameter is a resource load weight obtained by linearly weighting the resource utilization rate of each dimension of the use node; the stability parameter is a node stability value obtained according to the node online rate and the task completion rate;
respectively generating three sorted lists according to the resource usage balance parameter, the load balance parameter and the stability parameter of each node, taking the position values of the nodes in the three sorted lists as three-dimensional space coordinates, and selecting the node with the minimum Chebyshev distance to the original point to deploy a new container;
the method specifically comprises the following steps:
step 1) obtaining node diI is more than or equal to 1 and less than or equal to K, and calculating the resource utilization rate variance of each node; the nodes are arranged in an ascending order according to the variance of the resource utilization rate to obtain a sequence L1；
Step 2) calculating the resource load weight of each node according to the resource utilization rate of each dimension obtained in the step 1); according to the resource load weight, the nodes are arranged in an ascending order to obtain a sequence L2；
Step 3) calculating a stable value of each node according to the historical record of each node; according to the stable value, the nodes are arranged in ascending order to obtain a sequence L3；
Step 4) three sequences L according to nodes1、L2And L3The position value in (3) maps the node to a three-dimensional space, and the Chebyshev distance dis value of the node is calculated:
and 5) selecting the node with the minimum dis value as an optimal node, and deploying the task to the node.
2. The task scheduling method applicable to the embedded container cluster according to claim 1, wherein the step 1) specifically comprises:
step 1-1) computing node diMemory usage rate M (d)i)：
M(di)＝((MenTotal-MemFree)/MemTotal)×100％
Wherein, MemTotal is the total memory amount, and MemFree is the memory margin;
step 1-2) computing node diCpu usage of C (d)i)：
C(di)＝((us+sy)/(us+sy+id))×100％
Wherein us is a user state, sy is a kernel state, and id is an idle state;
step 1-3) calculating node diNetwork load usage rate of N (d)i)：
N(di)＝((N1+N2)/M)×100％
Wherein, N1 is the incoming flow, N2 is the outgoing flow, and M is the maximum network throughput;
step 1-4) calculating node diAverage resource usage of (2):
wherein N is 3; r is1i＝M(di)，r2i＝C(di)，r3i＝N(di)；
Step 1-5) calculating the node d according to the average resource utilization rateiResource usage variance of (2):
step 1-6) according to the variance S of the resource utilization ratei 2Carrying out ascending arrangement on the K nodes to obtain a sequence L1。
3. The task scheduling method applicable to the embedded container cluster according to claim 2, wherein the step 2) specifically comprises:
step 2-1) obtaining the memory utilization rate M (d)i) Cpu utilization C (d)i) And network load usage rate N (d)i) Weight k ofj，1≤j≤N；
Step 2-2 computing node diResource load weight W (d)i)：
Wherein, γjTo adjust the coefficients of the weights, and said gammajGreater than 1, R1(di)＝M(di)，R2(di)＝C(di)，R3(di)＝N(di) (ii) a p is a threshold;
step 2-3) according to the resource load weight W (d)i) Carrying out ascending arrangement on the K nodes to obtain a sequence L2。
4. The task scheduling method applicable to the embedded container cluster according to claim 3, wherein the step 1) specifically comprises:
step 3-1) calculating node diA stable value of Z (d)i)：
Wherein M is the number of unit time slices; n isjNumber of tasks completed in a unit time slice for a node, NjReceiving the total number of tasks, t, for a unit time slice of a nodejIs the on-line time length in the unit time slice of the node, tpIs a unit time slice;
step 3-2) based on the stabilized value Z (d)i) Carrying out ascending arrangement on the K nodes to obtain a sequence L3。
5. The task scheduling method applicable to the embedded container cluster according to claim 4, wherein the step 4) is specifically:
calculating the Chebyshev distance dis from the node to the node in the three-dimensional space according to the following formula:
wherein, ci1,ci2,ci3Are respectively node diIn the sequence L1，L2，L3The position value of (1).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810457653.0A CN110489200B (en) | 2018-05-14 | 2018-05-14 | Task scheduling method suitable for embedded container cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810457653.0A CN110489200B (en) | 2018-05-14 | 2018-05-14 | Task scheduling method suitable for embedded container cluster |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110489200A CN110489200A (en) | 2019-11-22 |
| CN110489200B true CN110489200B (en) | 2022-03-08 |
Family
ID=68544858
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810457653.0A Active CN110489200B (en) | 2018-05-14 | 2018-05-14 | Task scheduling method suitable for embedded container cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110489200B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111046102B (en) * | 2019-11-27 | 2023-10-31 | 复旦大学 | High performance blockchain service system |
| US20210191756A1 (en) * | 2019-12-19 | 2021-06-24 | Huawei Technologies Co., Ltd. | Methods and apparatus for resource scheduling of resource nodes of a computing cluster or a cloud computing platform |
| CN112019628A (en) * | 2020-09-01 | 2020-12-01 | 江西凌峰售电有限公司 | Low-delay low-energy-consumption intelligent platform system based on Internet of things |
| CN112114950A (en) * | 2020-09-21 | 2020-12-22 | 中国建设银行股份有限公司 | Task scheduling method and device and cluster management system |
| CN113377495B (en) * | 2021-05-17 | 2024-02-27 | 杭州中港科技有限公司 | Dock cluster deployment optimization method based on heuristic ant colony algorithm |
| US20220398134A1 (en) * | 2021-06-11 | 2022-12-15 | International Business Machines Corporation | Allocation of services to containers |
| CN116132447A (en) * | 2022-12-21 | 2023-05-16 | 天翼云科技有限公司 | Load balancing method and device based on Kubernetes |
| CN117076142B (en) * | 2023-10-17 | 2024-01-30 | 阿里云计算有限公司 | Multi-tenant resource pool configuration method and multi-tenant service system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103207804A (en) * | 2013-04-07 | 2013-07-17 | 杭州电子科技大学 | MapReduce load simulation method based on cluster job logging |
| CN105607947A (en) * | 2015-12-11 | 2016-05-25 | 西北工业大学 | Novel cloud environment virtual machine scheduling method |
| CN106776005A (en) * | 2016-11-23 | 2017-05-31 | 华中科技大学 | A kind of resource management system and method towards containerization application |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006092053A (en) * | 2004-09-22 | 2006-04-06 | Nec Corp | System use ratio management device, and system use ratio management method to be used for the same device and its program |
| CN103246550A (en) * | 2012-02-09 | 2013-08-14 | 深圳市腾讯计算机系统有限公司 | Multitask dispatching method and system based on capacity |
-
2018
- 2018-05-14 CN CN201810457653.0A patent/CN110489200B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103207804A (en) * | 2013-04-07 | 2013-07-17 | 杭州电子科技大学 | MapReduce load simulation method based on cluster job logging |
| CN105607947A (en) * | 2015-12-11 | 2016-05-25 | 西北工业大学 | Novel cloud environment virtual machine scheduling method |
| CN106776005A (en) * | 2016-11-23 | 2017-05-31 | 华中科技大学 | A kind of resource management system and method towards containerization application |
Non-Patent Citations (1)
| Title |
|---|
| 基于Docker Swarm集群的动态加权调度策略;黄凯;《计算机应用》;20180510;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110489200A (en) | 2019-11-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110489200B (en) | Task scheduling method suitable for embedded container cluster | |
| US11010188B1 (en) | Simulated data object storage using on-demand computation of data objects | |
| US9910888B2 (en) | Map-reduce job virtualization | |
| CN110297699B (en) | Scheduling method, scheduler, storage medium and system | |
| WO2019192263A1 (en) | Task assigning method, apparatus and device | |
| WO2011110026A1 (en) | Method and apparatus for realizing load balance of resources in data center | |
| CN109067834B (en) | Discrete particle swarm scheduling algorithm based on oscillation type inertia weight | |
| Zhang et al. | Heterogeneity aware dominant resource assistant heuristics for virtual machine consolidation | |
| Tseng et al. | Link-aware virtual machine placement for cloud services based on service-oriented architecture | |
| JP6129290B1 (en) | Method and system for recommending application parameter settings and system specification settings in distributed computing | |
| CN109361750A (en) | Resource allocation methods, device, electronic equipment, storage medium | |
| CN110597598B (en) | Control method for virtual machine migration in cloud environment | |
| CN108845886A (en) | Cloud computing energy consumption optimization method and system based on phase space | |
| CN106502761B (en) | Virtual machine deployment method capable of efficiently utilizing resources | |
| Huang et al. | Fuzzy clustering with feature weight preferences for load balancing in cloud | |
| CN110308965B (en) | Rule-based heuristic virtual machine distribution method and system for cloud data center | |
| CN109041236A (en) | A kind of wireless resource allocation methods and device of difference weight business | |
| Zhiyong et al. | An improved container cloud resource scheduling strategy | |
| Li et al. | On scheduling of high-throughput scientific workflows under budget constraints in multi-cloud environments | |
| CN116166181A (en) | Cloud monitoring method and cloud management platform | |
| Chen et al. | A Cloud Task Scheduling Algorithm Based on Users' Satisfaction | |
| US20210185119A1 (en) | A Decentralized Load-Balancing Method for Resource/Traffic Distribution | |
| CN112579246A (en) | Virtual machine migration processing method and device | |
| Peng et al. | The realization of load balancing algorithm in cloud computing | |
| CN115361284B (en) | Deployment adjustment method of virtual network function based on SDN |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210803 Address after: Room 1601, 16th floor, East Tower, Ximei building, No. 6, Changchun Road, high tech Industrial Development Zone, Zhengzhou, Henan 450001 Applicant after: Zhengzhou xinrand Network Technology Co.,Ltd. Address before: 100190, No. 21 West Fourth Ring Road, Beijing, Haidian District Applicant before: INSTITUTE OF ACOUSTICS, CHINESE ACADEMY OF SCIENCES |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |