CN110348681B - Power CPS dynamic load distribution method - Google Patents
Power CPS dynamic load distribution method Download PDFInfo
- Publication number
- CN110348681B CN110348681B CN201910481676.XA CN201910481676A CN110348681B CN 110348681 B CN110348681 B CN 110348681B CN 201910481676 A CN201910481676 A CN 201910481676A CN 110348681 B CN110348681 B CN 110348681B
- Authority
- CN
- China
- Prior art keywords
- task
- server node
- server
- bqe
- node
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention relates to the technical field of big data, in particular to a dynamic load distribution method of a power CPS (control system), which comprises the following steps: m1, establishing a distance table between servers, a task data queue table and a task allocation node table based on data transmission time; m2, generating a new task and inquiring the node state, and starting a task transfer strategy when the node is overloaded or overloaded; m3, querying the residual nodes in the priority allocation queue list, and if the residual nodes exist in the priority allocation queue list, transferring the task in the same order; m4, using the new task node j as a sending node, inquiring the task cost number in the node queue, M5, taking the node with the minimum Ci as a transfer node, and transferring the task from the new task node. By using the dynamic load distribution method of the electric power CPS, the task processing efficiency is accelerated by dynamically adjusting the task sequence of the server, and the overall pressure of the server cluster is reduced.
Description
Technical Field
The invention relates to the technical field of big data, in particular to a dynamic load distribution method for a power CPS.
Background
A Cyber Physical System (CPS) is a multidimensional complex System that integrates computing, network and Physical environment into a whole by 3C (Communication, Control) technology, and realizes real-time sensing, dynamic Control and information service of a large-scale engineering System by multi-technology organic fusion. Potential applications of the power CPS: analyzing the influence on the safety and the economy of the power system; and managing and controlling the load. The development of smart grids depends to a large extent on the development of information technology and distributed parallel technology. The deployment of the electric power CPS cannot avoid Cloud Computing, and the Cloud Computing (Cloud Computing) represents the development front of a fine-granular distributed parallel technology, is a brand-new Computing mode which is rapidly developed in recent years and is a general name of a plurality of new Computing technologies; it represents a large-scale, distributed computing model based on the Internet. Often, a server cluster needs to deploy a corresponding algorithm flow for a new calculation rule.
The invention discloses a security analysis method of an electric power CPS (control system performance) based on monitoring big data mining, which is disclosed by China published patent No. CN 109299160A on 2019, 2 month and 1 day, and relates to the technical field of ECPS security analysis. The method comprises the steps of firstly, establishing a general architecture for big data analysis of a dispatching control system, and mining a high-risk equipment set by taking an equipment risk value as a target; combining the CPS concept with the characteristics of the power system, and establishing a steady-state and dynamic model of the power information system; and aiming at each high-risk equipment set, evaluating whether the communication network is blocked or not by using a dynamic model of the electric power information system, calculating a time-varying path of the performance index of the information system in a later period, judging the electric power equipment which is possibly out of control based on the time-varying path, and giving an alarm on a data platform of a regulation and control center. The technical scheme makes up the defect of insufficient utilization of mass data in the prior art.
The method calls the historical data to calculate the dynamic model of the existing computing service, and the essence of the method is that the iterative algorithm or the genetic algorithm is used for learning and counting the historical data, a large amount of server resources are occupied, the model needs to be continuously updated after new data is established, the server resources need to be continuously occupied, the system establishment period is greatly prolonged when the method is used in a limited server resource environment, and the allocation of the server resources by the method depends on the data calculation load distribution history in the learning process and a data reading limit mode is set to avoid server blockage or server overloading.
Disclosure of Invention
The invention provides a dynamic load distribution method of an electric power CPS (control performance server) aiming at the defect that the conventional electric power CPS system only depends on a centralized allocation or queue transmission mode, and the task processing efficiency is accelerated by dynamically adjusting the task sequence of a server so as to reduce the overall pressure of a server cluster.
In order to solve the technical problems, the inventor adopts the following technical scheme: a power CPS dynamic load distribution method comprises the following steps: m1, establishing a distance table D between server nodes represented by time required for data transmission, acquiring a task data queue (b1, b2, … …, bm), wherein M is the total number of tasks, and establishing a task distribution state table [ qe1, qe2, … …, qen)]Qei, representing the current task amount distributed by the server i, n being the total number of the server nodes, and selecting a preferred distribution queue from the server node queues; m2, when a new task is generated, inquiring the current task allocation queue state, if the current server node is overloaded, overloaded or the number of continuously received new tasks exceeds a set threshold, entering a step M3 to execute a task transfer strategy, otherwise, repeating the step; m3, querying the remaining server nodes in the priority allocation queue, if the remaining server nodes exist in the priority allocation queue list, transferring the task to the server node, if the remaining server nodes do not exist in the priority allocation queue list, querying the remaining server nodes in the whole server node queue, if the remaining server nodes exist, transferring the task to the server node, otherwise, waiting for T time and then re-executing the step; m4, taking the new task server node j as the sending node, the task cost C of each server node in the server node queuei，
Wherein C isiTransferring the data amount required to generate a new task once from a server node j to a server node with an inquired code number i, wherein i is more than or equal to 1 and less than or equal to n, dijThe distance between the server node i and the server node j is defined, and B is the task data volume of the new task; m5, take CiAnd the minimum task load transfer strategy is used as a final power CPS dynamic load distribution strategy.
Preferably, step M2 includes the following sub-steps: a1, setting task transition strategy thresholds Q1 and Q2, wherein Q1< Q2; a2, inquiring the state of the server node, wherein the number of tasks of the current inquired server node is Q, and when Q is more than Q2, starting a task transfer strategy; a3, the current task number of the query server node is Q, when Q2> Q > Q1, the server node has no action; a4, the number of tasks of the current query server node is Q, and when Q is less than Q1, the server node is added into a priority distribution queue; a5, the task transfer strategy only transfers the newly generated task in the task queue of the query server node, and does not transfer the task in execution.
Preferably, the M1 comprises the following sub-steps: b1, calculating the data size Bqe of the newly generated task; b2, splitting the task package into new tasks in batch mode to make the new tasks become task queues [ Bqe1, Bqe2, Bqe3, … …, Bqen ]](ii) a B3, re-executing the step M2 by the task queue of the step B2, and transferring the tasks in sequence; b4, recording the status of all server nodes not overloaded or overloaded after i tasks are transferred [ Bqe1, Bqe2, Bqe3, … …, Bqei ] statistics]Total amount of BqezAnd recording; b5, data volume Bqe of new task generated subsequently is directly connected with BqezBy comparison, greater than BqezIs less than Bqe, then step B2 is performedzThen step M2 is entered directly.
Preferably, if the amount of new task data generated is less than BqezAnd the server status is overloaded or overloaded, steps B2 through B4 are re-executed after step B4, and the server status is updated BqezThe numerical value of (c).
By splitting the sub-packets of the comprehensive tasks, the tasks can be distributed to other servers in the server cluster to accelerate the processing of one comprehensive task, and simultaneously the load pressure of a single server is reduced, while the iterative judgment of the step B5 is simple judgment of the data volume of the tasks, the confirmation of the task state is accelerated through the judgment of the data volume, and the problem that the communication is occupied because the sub-packets are sent to other servers is reduced; and after a number of iterations this value will result in a value that is appropriate for the local server, with each server Bqe in the server cluster having a different value.
Preferably, step M4 includes the following sub-steps: c1, calculating the total amount of the task queue data of the current server node i as Bi; c2, under the condition that no new task is added into the queue, the time for the current server node i to complete the task is ti; c3, traversing other server nodes requesting for one time, and recording the time of obtaining the response, wherein the shortest response time is recorded as min-dti; c4 according to the formula
Obtaining priority balance parameter C of server node isi(ii) a C5, according to C'i＝Ci-CsiCost of updating data volume Ci(ii) a C6, C 'as step C5'iAs a data volume cost CiAnd step M4 is carried in, and the selection of the server node for task data transfer is completed.
Compared with the cost value of a simple and rough computing task, after the concept of server processing cost is introduced, the task can be preferentially distributed to the servers with strong actual processing capacity when the task is transferred, and the possibility that the transferred task is processed instead of being transferred again is provided by accelerating the processing of the transferred task.
The method has the substantial effect that by using the dynamic load distribution method of the electric power CPS, the task processing efficiency is accelerated by dynamically adjusting the task sequence of the server, and the overall pressure of the server cluster is reduced.
Drawings
Fig. 1 is a flowchart of a dynamic load distribution method according to an embodiment.
Detailed Description
The technical solution of the present invention is further specifically described below by way of specific examples in conjunction with the accompanying drawings.
The first embodiment is as follows:
a power CPS dynamic load distribution method, as shown in fig. 1, comprising the steps of: m1, establishing a distance table D between server nodes represented by time required for data transmission, acquiring a task data queue (b1, b2, … …, bm), wherein M is the total number of tasks, and establishing a task distribution state table [ qe1, qe2, … …, qen)]Qei represents the current task amount allocated by the server i, n is the total number of server nodes, and the preferred allocation queue is selected from the server node queues. The method specifically comprises the following substeps: b1, calculating the data size Bqe of the newly generated task; b2, splitting the task package into new tasks in batch mode to make the new tasks become task queues [ Bqe1, Bqe2, Bqe3, … …, Bqen ]](ii) a B3, re-executing the step M2 by the task queue of the step B2, and transferring the tasks in sequence; b4, recording the status of all server nodes not overloaded or overloaded after i tasks are transferred [ Bqe1, Bqe2, Bqe3, … …, Bqei ] statistics]Total amount of BqezAnd recording; b5, data volume Bqe of new task generated subsequently is directly connected with BqezBy comparison, greater than BqezIs less than Bqe, then step B2 is performedzThen step M2 is entered directly. If the amount of the generated new task data is less than BqezAnd the server status is overloaded or overloaded, steps B2 through B4 are re-executed after step B4, and the server status is updated BqezThe numerical value of (c).
M2, when a new task is generated, the current task allocation queue state is inquired, if the current server node is overloaded, overloaded or the number of new tasks continuously received exceeds the set threshold, the step M3 is entered to execute the task transfer strategy, otherwise, the step is repeated. The method specifically comprises the following substeps: a1, setting task transition strategy thresholds Q1 and Q2, wherein Q1< Q2; a2, inquiring the state of the server node, wherein the number of tasks of the current inquired server node is Q, and when Q is more than Q2, starting a task transfer strategy; a3, the current task number of the query server node is Q, when Q2> Q > Q1, the server node has no action; a4, the number of tasks of the current query server node is Q, and when Q is less than Q1, the server node is added into a priority distribution queue; a5, the task transfer strategy only transfers the newly generated task in the task queue of the query server node, and does not transfer the task in execution.
M3, querying the remaining server nodes in the priority allocation queue, if the remaining server nodes exist in the priority allocation queue list, transferring the task to the server node, if the remaining server nodes do not exist in the priority allocation queue list, querying the remaining server nodes in the whole server node queue, if the remaining server nodes exist, transferring the task to the server node, otherwise, waiting for T time and then re-executing the step.
M4, taking the new task server node j as the sending node, the task cost C of each server node in the server node queuei，
Wherein C isiTransferring the data amount required to generate a new task once from a server node j to a server node with an inquired code number i, wherein i is more than or equal to 1 and less than or equal to n, dijThe distance between the server node i and the server node j is, and B is the task data volume of the new task. The method specifically comprises the following substeps: c1, calculating the total amount of the task queue data of the current server node i as Bi; c2, under the condition that no new task is added into the queue, the time for the current server node i to complete the task is ti; c3, traversing other server nodes requesting for one time, and recording the time of obtaining the response, wherein the shortest response time is recorded as min-dti; c4 according to the formula
Obtaining priority balance parameter C of server node isi(ii) a C5, according to C'i＝Ci-CsiCost of updating data volume Ci(ii) a C6, C 'as step C5'iAs a data volume cost CiAnd step M4 is carried in, and the selection of the server node for task data transfer is completed.
M5, take CiAnd the minimum task load transfer strategy is used as a final power CPS dynamic load distribution strategy.
By splitting the sub-packets of the comprehensive tasks, the tasks can be distributed to other servers in the server cluster to accelerate the processing of one comprehensive task, and simultaneously the load pressure of a single server is reduced, while the iterative judgment of the step B5 is simple judgment of the data volume of the tasks, the confirmation of the task state is accelerated through the judgment of the data volume, and the problem that the communication is occupied because the sub-packets are sent to other servers is reduced; and after a number of iterations this value will result in a value that is appropriate for the local server, with each server Bqe in the server cluster having a different value.
Compared with simple and rough computing task cost values, after the concept of server processing cost is introduced, the tasks can be preferentially distributed to the servers with strong actual processing capacity when being transferred, and the possibility that the transferred tasks are processed but not transferred again is provided by the processing goods of the transferred tasks.
The above-described embodiments are only preferred embodiments of the present invention, and are not intended to limit the present invention in any way, and other variations and modifications may be made without departing from the spirit of the invention as set forth in the claims.
Claims (4)
1. A power CPS dynamic load distribution method is characterized by comprising the following steps:
m1, establishing a distance table D between server nodes expressed by time required for data transmission, and acquiring a task data queue (b)1，b2，……，bm) And m is the total number of tasks, and a task allocation state table [ qe ] is established1，qe2，……，qen]，qeiRepresenting the current task amount distributed by the server i, wherein n is the total number of the server nodes, and selecting a preferred distribution queue from the server node queues;
m2, when a new task is generated, inquiring the current task allocation queue state, if the current server node is overloaded, overloaded or the number of continuously received new tasks exceeds a set threshold, entering a step M3 to execute a task transfer strategy, otherwise, repeating the step;
m3, querying the remaining server nodes in the priority allocation queue, if the remaining server nodes exist in the priority allocation queue list, transferring the task to the server node, if the remaining server nodes do not exist in the priority allocation queue list, querying the remaining server nodes in the whole server node queue, if the remaining server nodes exist, transferring the task to the server node, otherwise, waiting for T time and then re-executing the step;
m4, taking the new task server node j as a sending node, calculating the task cost C of each server node in the server node queuei，
Wherein C isiTransferring the data amount required to generate a new task once from a server node j to a server node with an inquired code number i, wherein i is more than or equal to 1 and less than or equal to n, dijThe distance between the server node i and the server node j is defined, and B is the task data volume of the new task;
step M4 includes the following sub-steps:
c1, calculating the total amount of the task queue data of the current server node i as Bi;
c2, estimating the time t for the current server node i to complete the task under the condition that no new task is added into the queuei；
C3, traversing other server nodes requesting once, and recording the time of getting response, wherein the shortest response time is recorded as min-dti；
C4 according to the formula
Obtaining priority balance parameter C of server node isi；
C5, according to C'i＝Ci-CsiCost of updating data volume Ci；
C6, C 'as step C5'iAs a data volume cost CiCarrying out a step M4, and completing the selection of the server node for task data transfer; m5, take CiAnd the minimum task load transfer strategy is used as a final power CPS dynamic load distribution strategy.
2. A power CPS dynamic load distribution method as claimed in claim 1, wherein step M2 comprises the following sub-steps:
a1, setting task transition strategy thresholds Q1 and Q2, wherein Q1< Q2;
a2, inquiring the state of the server node, wherein the number of tasks of the current inquired server node is Q, and when Q is more than Q2, starting a task transfer strategy;
a3, the current task number of the query server node is Q, when Q2> Q > Q1, the server node has no action;
a4, the number of tasks of the current query server node is Q, and when Q is less than Q1, the server node is added into a priority distribution queue;
a5, the task transfer strategy only transfers the newly generated task in the task queue of the query server node, and does not transfer the task in execution.
3. A power CPS dynamic load distribution method according to claim 1 or 2, wherein step M1 comprises the following sub-steps:
b1, calculating the data size Bqe of the newly generated task;
b2, splitting the task package into new tasks in batch processing to make the new tasks become task queue [ Bqe ]1，Bqe2，Bqe3，……，Bqen]；
B3, re-executing the step M2 by the task queue of the step B2, and transferring the tasks in sequence;
b4, recording the status of all server nodes not overloaded or overloaded after i tasks are transferred [ Bqe ] for statistics1，Bqe2，Bqe3，……，Bqei]Total amount of BqezAnd recording;
b5, data volume Bqe of new task generated subsequently is directly connected with BqezBy comparison, greater than BqezIs less than Bqe, then step B2 is performedzThen step M2 is entered directly.
4. The method for dynamic load distribution for electric power CPS as claimed in claim 3, wherein if the amount of generated new task data is less than BqezAnd the server status is overloaded or overloaded, steps B2 through B4 are re-executed after step B4, and the server status is updated BqezThe numerical value of (c).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910481676.XA CN110348681B (en) | 2019-06-04 | 2019-06-04 | Power CPS dynamic load distribution method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910481676.XA CN110348681B (en) | 2019-06-04 | 2019-06-04 | Power CPS dynamic load distribution method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110348681A CN110348681A (en) | 2019-10-18 |
| CN110348681B true CN110348681B (en) | 2022-02-18 |
Family
ID=68181494
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910481676.XA Active CN110348681B (en) | 2019-06-04 | 2019-06-04 | Power CPS dynamic load distribution method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110348681B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110737695B (en) * | 2019-10-08 | 2020-11-24 | 重庆紫光华山智安科技有限公司 | Multistage data retrieval optimization method based on terminal computing power and dynamic empowerment |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09160884A (en) * | 1995-12-04 | 1997-06-20 | Mitsubishi Electric Corp | Dynamic load-distributed parallel computer |
| CN101014002A (en) * | 2006-12-12 | 2007-08-08 | 华为技术有限公司 | Cluster message transmitting method and distributed cluster system |
| CN102708011A (en) * | 2012-05-11 | 2012-10-03 | 南京邮电大学 | Multistage load estimating method facing task scheduling of cloud computing platform |
| CN103077082A (en) * | 2013-01-08 | 2013-05-01 | 中国科学院深圳先进技术研究院 | Method and system for distributing data center load and saving energy during virtual machine migration |
| CN103401947A (en) * | 2013-08-20 | 2013-11-20 | 曙光信息产业(北京)有限公司 | Method and device for allocating tasks to multiple servers |
| CN103683337A (en) * | 2013-12-05 | 2014-03-26 | 华南理工大学 | Interconnected power system CPS instruction dynamic allocation and optimization method |
| CN104539551A (en) * | 2014-12-22 | 2015-04-22 | 清华大学 | Virtual service migration method for routing and switching platform |
| CN104994156A (en) * | 2015-07-01 | 2015-10-21 | 北京京东尚科信息技术有限公司 | Load balancing method and system for cluster |
| CN105007312A (en) * | 2015-07-03 | 2015-10-28 | 叶秀兰 | Method and system for controlling adaptive load-balancing of cloud computing server |
| CN105516328A (en) * | 2015-12-18 | 2016-04-20 | 浪潮(北京)电子信息产业有限公司 | Dynamic load balancing method and system, and devices used for distributed storage system |
| CN105516360A (en) * | 2016-01-19 | 2016-04-20 | 苏州帕科泰克物联技术有限公司 | Method and device for load balance of computer |
| CN105718312A (en) * | 2016-01-20 | 2016-06-29 | 华南理工大学 | Multi-queue back-filling job scheduling method oriented to living organism gene sequencing calculation task |
| CN105959411A (en) * | 2016-06-30 | 2016-09-21 | 中原智慧城市设计研究院有限公司 | Dynamic load balance distributed processing method in cloud computing environment based on coordination |
| CN106844051A (en) * | 2017-01-19 | 2017-06-13 | 河海大学 | The loading commissions migration algorithm of optimised power consumption in a kind of edge calculations environment |
| CN107196865A (en) * | 2017-06-08 | 2017-09-22 | 中国民航大学 | A kind of adaptive threshold overload moving method of Load-aware |
| CN107580023A (en) * | 2017-08-04 | 2018-01-12 | 山东大学 | A kind of the stream process job scheduling method and system of dynamic adjustment task distribution |
| CN108429815A (en) * | 2018-03-23 | 2018-08-21 | 重庆邮电大学 | Dynamic resource scheduling method based on OpenStack |
| CN108446180A (en) * | 2018-03-23 | 2018-08-24 | 南京航空航天大学 | A kind of data center dynamic method for scheduling task based on Data Migration |
-
2019
- 2019-06-04 CN CN201910481676.XA patent/CN110348681B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09160884A (en) * | 1995-12-04 | 1997-06-20 | Mitsubishi Electric Corp | Dynamic load-distributed parallel computer |
| CN101014002A (en) * | 2006-12-12 | 2007-08-08 | 华为技术有限公司 | Cluster message transmitting method and distributed cluster system |
| CN102708011A (en) * | 2012-05-11 | 2012-10-03 | 南京邮电大学 | Multistage load estimating method facing task scheduling of cloud computing platform |
| CN103077082A (en) * | 2013-01-08 | 2013-05-01 | 中国科学院深圳先进技术研究院 | Method and system for distributing data center load and saving energy during virtual machine migration |
| CN103401947A (en) * | 2013-08-20 | 2013-11-20 | 曙光信息产业(北京)有限公司 | Method and device for allocating tasks to multiple servers |
| CN103683337A (en) * | 2013-12-05 | 2014-03-26 | 华南理工大学 | Interconnected power system CPS instruction dynamic allocation and optimization method |
| CN104539551A (en) * | 2014-12-22 | 2015-04-22 | 清华大学 | Virtual service migration method for routing and switching platform |
| CN104994156A (en) * | 2015-07-01 | 2015-10-21 | 北京京东尚科信息技术有限公司 | Load balancing method and system for cluster |
| CN105007312A (en) * | 2015-07-03 | 2015-10-28 | 叶秀兰 | Method and system for controlling adaptive load-balancing of cloud computing server |
| CN105516328A (en) * | 2015-12-18 | 2016-04-20 | 浪潮(北京)电子信息产业有限公司 | Dynamic load balancing method and system, and devices used for distributed storage system |
| CN105516360A (en) * | 2016-01-19 | 2016-04-20 | 苏州帕科泰克物联技术有限公司 | Method and device for load balance of computer |
| CN105718312A (en) * | 2016-01-20 | 2016-06-29 | 华南理工大学 | Multi-queue back-filling job scheduling method oriented to living organism gene sequencing calculation task |
| CN105959411A (en) * | 2016-06-30 | 2016-09-21 | 中原智慧城市设计研究院有限公司 | Dynamic load balance distributed processing method in cloud computing environment based on coordination |
| CN106844051A (en) * | 2017-01-19 | 2017-06-13 | 河海大学 | The loading commissions migration algorithm of optimised power consumption in a kind of edge calculations environment |
| CN107196865A (en) * | 2017-06-08 | 2017-09-22 | 中国民航大学 | A kind of adaptive threshold overload moving method of Load-aware |
| CN107580023A (en) * | 2017-08-04 | 2018-01-12 | 山东大学 | A kind of the stream process job scheduling method and system of dynamic adjustment task distribution |
| CN108429815A (en) * | 2018-03-23 | 2018-08-21 | 重庆邮电大学 | Dynamic resource scheduling method based on OpenStack |
| CN108446180A (en) * | 2018-03-23 | 2018-08-24 | 南京航空航天大学 | A kind of data center dynamic method for scheduling task based on Data Migration |
Non-Patent Citations (2)
| Title |
|---|
| 一种面向分布式服务器集群的动态负载均衡系统的实现;丁学智;《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》;20150415;第2章 * |
| 基于动态阈值分配的流媒体边缘云会话迁移策略;姜同全等;《计算机工程》;20170131;第55-60页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110348681A (en) | 2019-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104969213B (en) | Data flow for low latency data access is split | |
| CN102299959B (en) | Load balance realizing method of database cluster system and device | |
| CN108076158B (en) | Minimum load route selection method and system based on naive Bayes classifier | |
| US20220417156A1 (en) | Network burst load evacuation method for edge servers | |
| CN102859961B (en) | There is the distributed video transcoding system of adaptive file process | |
| CN107872402A (en) | The method, apparatus and electronic equipment of global traffic scheduling | |
| CN103986766A (en) | Self-adaptation load balancing job task scheduling method and device | |
| WO2023066084A1 (en) | Computing power distribution method and apparatus, and computing power server | |
| CN108491255B (en) | Self-service MapReduce data optimal distribution method and system | |
| CN114039918A (en) | Information age optimization method and device, computer equipment and storage medium | |
| CN103942197A (en) | Data monitoring processing method and device | |
| CN113132490A (en) | MQTT protocol QoS mechanism selection scheme based on reinforcement learning | |
| CN112187891A (en) | Load optimization method and device of edge computing node set based on multiple services | |
| CN105119751A (en) | Service evaluation and selection method based on environment real-time perceiving | |
| CN116501711A (en) | Computing power network task scheduling method based on 'memory computing separation' architecture | |
| CN112261120A (en) | Cloud-side cooperative task unloading method and device for power distribution internet of things | |
| CN110348681B (en) | Power CPS dynamic load distribution method | |
| CN106506647A (en) | A kind of client has the intelligence community cloud storage system of data backup device | |
| Huang | The value-of-information in matching with queues | |
| CN109376509A (en) | A kind of KVM task distributes system and automatic distributing method automatically | |
| Zhou et al. | Study on the evolutionary optimisation of the topology of network control systems | |
| CN115562861A (en) | Method and apparatus for data processing for data skew | |
| CN114938376A (en) | Industrial Internet of things based on priority processing data and control method thereof | |
| CN112165721B (en) | Multi-service task unloading and service migration method based on edge computing | |
| CN106357676A (en) | Method for optimizing overhead of cloud service resource |
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 |