CN110348681A - A kind of electric power CPS dynamic load distribution method - Google Patents
A kind of electric power CPS dynamic load distribution method Download PDFInfo
- Publication number
- CN110348681A CN110348681A CN201910481676.XA CN201910481676A CN110348681A CN 110348681 A CN110348681 A CN 110348681A CN 201910481676 A CN201910481676 A CN 201910481676A CN 110348681 A CN110348681 A CN 110348681A
- Authority
- CN
- China
- Prior art keywords
- task
- server node
- queue
- node
- bqe
- 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
- 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 present invention relates to big data technical fields, and in particular to a kind of electric power CPS dynamic load distribution method, comprising: M1 is established between the server based on transmitting data time apart from table, task data queue table, task allocation node table；M2, new task generates and query node state, when node heavy duty or overload then start task transition strategy；M3 inquires remaining node in preferential distribution queue table, and cis-position shifts task if preferential distribution queue table has remaining node；M4, using new task node j as sending node, task cost number, M5 in query node queue, taking the smallest node of Ci is transfering node, shifts task from new task node.Using electric power CPS dynamic load distribution method of the invention, server task sequence is adjusted by dynamic and mitigates server cluster integral pressure to accelerate task treatment effeciency.
Description
Technical field
The present invention relates to big data technical fields, and in particular to a kind of electric power CPS dynamic load distribution method.
Background technique
Information physical emerging system (Cyber Physical System, be abbreviated as CPS) is to pass through 3C
(Computation, Communication, Control) technology answers the multidimensional that calculating, network and physical environment combine together
Miscellaneous system is organically blended by more technologies, realizes real-time perception, dynamic control and the information service of heavy construction system.Electric power
The potential application of CPS: the impact analysis to power system security and economy；Load is managed and is controlled.Intelligence electricity
The development of net depends greatly on the development of information technology and distributed parallel technology.And wherein to the portion of electric power CPS
Unavoidable cloud computing is affixed one's name to, before cloud computing (Cloud Computing) represents the development of fine particulate distributed parallel technology
Edge is a kind of brand-new calculating mode being rapidly developed in recent years, is the general designation of several new computing techniques；Which represent one
Large-scale distributed calculating mode of the kind based on Internet.Server cluster is generally required to new computation rule deployment pair
The algorithm flow answered.
Chinese 109299160 A of publication CN, publication date on 2 1st, 2019, denomination of invention one kind was based on monitoring
The electric power CPS Safety Analysis Method that big data is excavated discloses a kind of electric power CPS safety excavated based on monitoring big data
Analysis method is related to ECPS safety analysis technical field.This method initially sets up the total of Dispatching Control System big data analysis
Body framework carries out high risk equipment collection as target using equipment Risk value and excavates；By CPS concept in conjunction with electric system feature, build
The stable state and dynamic model of vertical power information system；For each high risk equipment collection, power information system dynamic analog is utilized
Type assessment communication network in whether can block, and in the latter period of calculating the performance indicator of information system time-varying road
Diameter may lost the power equipment of control based on this judgement, and provide alarm in regulation centre data platform.Its technical solution is more
It has mended the prior art and insufficient deficiency is utilized to mass data.
But its shortcoming be the invention have invoked historical data calculate it is existing calculate service dynamic model,
Matter is to be counted with Class of Iterative algorithm or hereditary class algorithm to the study of historical data, needs to occupy a large amount of server resource, and
And need persistently model to be updated to appoint after establishing new data and so need persistently to occupy server resource, in limited clothes
So using then establishing a large amount of extension systems the period in business device resource environment, and the method is deployed server resource and is relied on
Data calculated load allocation history in learning process, and setting read in the mode of data quota to avoid server from blocking or take
Load is thought highly of in business.
Summary of the invention
The deficiency for concentrating the mode of allotment or queue conveying is relied solely on the present invention be directed to existing electric power CPS system, is mentioned
A kind of electric power CPS dynamic load distribution method is gone out, server task sequence is adjusted by dynamic to accelerate task treatment effeciency
Mitigate server cluster integral pressure.
In order to solve the above technical problems, the present inventor uses a kind of following technical scheme: the dynamic load distribution side electric power CPS
Method, comprising: M1 is established to transmit between the server node indicated the time required to data apart from table D, acquisition task data queue
(b1, b2 ... ..., bm), m are total task number amount, establish task distribution state table [qe1, qe2 ... ..., qen], and qei indicates clothes
Business device i assigned existing task amount, n are server node sum, and preferred allocation queue is selected from server node queue；
M2 inquires current task distribution queue state when new task generates, if the heavy duty of current server node, overload or continuous received
New task quantity is more than given threshold, then enters step M3 and execute task transition strategy, conversely, repeating this step；M3 is inquired excellent
Remaining server node in first distribution queue shifts task to the clothes if preferential distribution queue table has remaining server node
Business device node, then inquires the remaining server node of whole server node queues if it does not exist, such as there is remaining server section
Point then shifts task to the server node, conversely, re-executing this step after then waiting T time；M4, with new task server
Node j is sending node, the task cost C of each server node in server node queuei,Wherein CiIt is primary newly appointed to the server node transfer that code name is i is queried for server node j
The data volume generated required for business, wherein 1≤i≤n, dijIt is new for the distance between server node i and server node j, B
The task data amount of task；M5 takes CiThe smallest task load transition strategy, as final electric power CPS dynamic load distribution plan
Slightly.
Preferably, step M2 includes following sub-step: A1, set task transition strategy threshold value Q1 and Q2, wherein Q1 <
Q2；A2, query service device node state, current queries server node number of tasks are Q, work as Q > Q2 and then start task transfer plan
Slightly；A3, current queries server node number of tasks are Q, work as Q2 > Q > Q1, then server node attonity；A4, current queries clothes
Business device node tasks number is Q, and as Q < Q1, then preferential distribution queue is added in server node；A5, task transition strategy are only shifted and are looked into
Newly generated task in server node task queue is ask, does not shift task in execution.
Preferably, the M1 includes following sub-step: B1, the data volume Bqe of newly generated task is calculated；B2 is right
Newly generated task carry out batch processing split task packet make new task become task queue [Bqe1, Bqe2, Bqe3 ... ...,
Bqen]；B3 re-execute the steps M2 with step B2 task queue, successively shifts task；B4, after record shifts i task,
So that whole server node states are non-heavily loaded or overload, the total amount of statistics [Bqe1, Bqe2, Bqe3 ... ..., Bqei] are
Bqez, and record；B5, the data volume Bqe of the new task of subsequent generation are directly and BqezCompare, is greater than BqezNew task then hold
Row step B2 is less than BqezNew task be then directly entered step M2.
Preferably, if the new task data volume generated is less than Bqez, and server state heavy duty or overload, then in step
B4 re-execute the steps B2 to B4 later, and updates BqezNumerical value.
It is split by the subpackage to comprehensive task, other servers that task can be distributed in server cluster make
One comprehensive task acceleration processing, while the load pressure of individual server is reduced, and the judgement of the iteration of step B5 is then
To the simple judgment of task data amount, accelerate the confirmation to task status by data volume judgement, reduces since subpackage is sent
The problem of occupying communication to other servers；And after successive ignition this value can obtain adaptation and locally take
The value of business device, each the server B qe value in server cluster is unequal.
Preferably, step M4 includes following sub-step: C1, the job queue data for calculating current server node i is total
Amount is Bi；C2, in the case that queue is added without new task in estimation, current server node i completes task time as ti；C3, time
The primary other server nodes of request are gone through, and record the time responded, wherein the shortest response time is denoted as min-dti；
C4, according to calculating formulaObtain the priority balance parameters C of server node isi；C5, according to C 'i=Ci-
CsiMore amount of new data cost Ci；C6, with the C ' of step C5iAs data volume cost CiIt brings step M4 into, completes task data transfer
Server node selection.
It is in office after the concept for introducing server process cost compared to than simple and crude calculating task cost value
Business can preferentially give actual treatment ability strong server when transfer, the processing goods for being transferred task is accelerated to provide
It is transferred a possibility that task is processed rather than transfer again.
Substantial effect of the invention is to adjust using electric power CPS dynamic load distribution method of the invention by dynamic
Whole server task sequence mitigates server cluster integral pressure to accelerate task treatment effeciency.
Detailed description of the invention
Fig. 1 is one dynamic load distribution method flow block diagram of embodiment.
Specific embodiment
Below by specific embodiment, and in conjunction with attached drawing, technical scheme of the present invention will be further explained in detail.
Embodiment one:
A kind of electric power CPS dynamic load distribution method, as shown in Figure 1, comprising the following steps: M1 is established to transmit needed for data
Apart from table D between the server node that time indicates, obtaining task data queue (b1, b2 ... ..., bm), m is total task number amount,
Task distribution state table [qe1, qe2 ... ..., qen] is established, qei indicates that the assigned existing task amount of server i, n are service
Device node total number selects preferred allocation queue from server node queue.Following sub-step: B1 is specifically included, new produce is calculated
The data volume Bqe of raw task；B2 carries out batch processing fractionation task packet to newly generated task and new task is made to become task team
It arranges [Bqe1, Bqe2, Bqe3 ... ..., Bqen]；B3 re-execute the steps M2 with step B2 task queue, successively shifts task；
B4, after record shifts i task, so that the non-heavy duty of whole server node states or overload, statistics [Bqe1, Bqe2,
Bqe3 ... ..., Bqei] total amount be Bqez, and record；B5, the data volume Bqe of the new task of subsequent generation are directly and BqezThan
Compared with greater than BqezNew task then follow the steps B2, be less than BqezNew task be then directly entered step M2.If the new post generated
Data volume of being engaged in is less than Bqez, and server state heavy duty or overload, then B2 to B4 is re-execute the steps after step B4, and more
New BqezNumerical value.
M2 inquires current task distribution queue state when new task generates, if current server node is heavily loaded, overloads or connects
Continue received new task quantity more than given threshold, then enters step M3 and execute task transition strategy, conversely, repeating this step.
Following sub-step: A1 is specifically included, sets task transition strategy threshold value Q1 and Q2, wherein Q1 < Q2；A2, query service device node
State, current queries server node number of tasks are Q, work as Q > Q2 and then start task transition strategy；A3, current queries server section
Point number of tasks is Q, works as Q2 > Q > Q1, then server node attonity；A4, current queries server node number of tasks be Q, when Q <
Q1, then preferential distribution queue is added in server node；A5, in task transition strategy polling time server node task queue
Newly generated task does not shift task in execution.
M3, inquires remaining server node in preferential distribution queue, and such as preferential distribution queue table has remaining server section
Point then shifts task to the server node, then inquires the remaining server node of whole server node queues if it does not exist,
Task is shifted to the server node, conversely, re-executing this step after then waiting T time if there is remaining server node
Suddenly.
M4, using new task server node j as sending node, the task of each server node in server node queue
Cost Ci,Wherein CiOne is shifted to the server node that code name is i is queried for server node j
The data volume generated required for secondary new task, wherein 1≤i≤n, dijBetween server node i and server node j away from
From B is the task data amount of new task.Following sub-step: C1 is specifically included, the task queue of current server node i is calculated
Total amount of data is Bi；C2, in the case that queue is added without new task in estimation, current server node i completes task time as ti；
C3, traversal requests primary other server nodes, and records the time responded, wherein the shortest response time is denoted as min-
dti；C4, according to calculating formulaObtain the priority balance parameters C of server node isi；C5, according to C 'i=
Ci-CsiMore amount of new data cost Ci；C6, with the C ' of step C5iAs data volume cost CiIt brings step M4 into, completes task data
The server node of transfer selects.
M5 takes CiThe smallest task load transition strategy, as final electric power CPS dynamic load distribution strategy.
It is split by the subpackage to comprehensive task, other servers that task can be distributed in server cluster make
One comprehensive task acceleration processing, while the load pressure of individual server is reduced, and the judgement of the iteration of step B5 is then
To the simple judgment of task data amount, accelerate the confirmation to task status by data volume judgement, reduces since subpackage is sent
The problem of occupying communication to other servers；And after successive ignition this value can obtain adaptation and locally take
The value of business device, each the server B qe value in server cluster is unequal.
Compare simple and crude calculating task cost value, after the concept for introducing server process cost, task into
The strong server of actual treatment ability can be preferentially given when row transfer, the processing goods for being transferred task offer is accelerated to be turned
A possibility that shifting task is processed rather than transfer again.
Above-mentioned embodiment is only a preferred solution of the present invention, not the present invention is made in any form
Limitation, there are also other variations and modifications on the premise of not exceeding the technical scheme recorded in the claims.
Claims (7)
1. a kind of electric power CPS dynamic load distribution method characterized by comprising
M1 is established to transmit between the server node indicated the time required to data apart from table D, acquisition task data queue (b1,
b2... ..., bm), m is total task number amount, establishes task distribution state table [qe1, qe2... ..., qen], qeiIndicate server i quilt
The existing task amount of distribution, n are server node sum, and preferred allocation queue is selected from server node queue；
M2 inquires current task distribution queue state when new task generates, if current server node is heavily loaded, overloads or continuously connects
The new task quantity of receipts is more than given threshold, then enters step M3 and execute task transition strategy, conversely, repeating this step；
M3 inquires remaining server node in preferential distribution queue, if preferential distribution queue table has remaining server node
Transfer task is then inquired the remaining server node of whole server node queues if it does not exist, is such as deposited to the server node
Task is then shifted to the server node, conversely, re-executing this step after then waiting T time in remaining server node；
M4, using new task server node j as sending node, the task of each server node in calculation server node queue
Cost Ci,Wherein CiIt is shifted for server node j to the server node that code name is i is queried
The data volume generated required for new task, wherein 1≤i≤n, dijBetween server node i and server node j away from
From B is the task data amount of new task；
M5 takes CiThe smallest task load transition strategy, as final electric power CPS dynamic load distribution strategy.
2. a kind of electric power CPS dynamic load distribution method according to claim 1, which is characterized in that step M2 include with
Lower sub-step:
A1 sets task transition strategy threshold value Q1 and Q2, wherein Q1 < Q2；
A2, query service device node state, current queries server node number of tasks are Q, work as Q > Q2 and then start task transfer plan
Slightly；
A3, current queries server node number of tasks are Q, work as Q2 > Q > Q1, then server node attonity；
A4, current queries server node number of tasks are Q, and as Q < Q1, then preferential distribution queue is added in server node；
A5, newly generated task in task transition strategy polling time server node task queue, is not shifted in execution
Task.
3. a kind of electric power CPS dynamic load distribution method according to claim 1 or 2, which is characterized in that step M1 includes
Following sub-step:
B1 calculates the data volume Bqe of newly generated task；
B2 carries out batch processing fractionation task packet to newly generated task and new task is made to become task queue [Bqe1, Bqe2,
Bqe3... ..., Bqen]；
B3 re-execute the steps M2 with step B2 task queue, successively shifts task；
B4, after record shifts i task, so that the non-heavy duty of whole server node states or overload, count [Bqe1,
Bqe2, Bqe3... ..., Bqei] total amount be Bqez, and record；
B5, the data volume Bqe of the new task of subsequent generation are directly and BqezCompare, is greater than BqezNew task then follow the steps B2,
Less than BqezNew task be then directly entered step M2.
4. a kind of electric power CPS dynamic load distribution method according to claim 3, which is characterized in that if the new post generated
Data volume of being engaged in is less than Bqez, and server state heavy duty or overload, then B2 to B4 is re-execute the steps after step B4, and more
New BqezNumerical value.
5. a kind of electric power CPS dynamic load distribution method according to claim 1 or 2, which is characterized in that step M4 includes
Following sub-step:
C1, the job queue data total amount for calculating current server node i is Bi；
C2, in the case that queue is added without new task in estimation, current server node i completes task time as ti；
C3, traversal requests primary other server nodes, and records the time responded, wherein the shortest response time is denoted as
min-dti；
C4, according to calculating formulaObtain the priority balance parameters C of server node isi；
C5, according to C 'i=Ci-CsiMore amount of new data cost Ci；
C6, with the C ' of step C5iAs data volume cost CiIt brings step M4 into, completes the server node choosing of task data transfer
It selects.
6. a kind of electric power CPS dynamic load distribution method according to claim 3, which is characterized in that step M4 include with
Lower sub-step:
C1, the job queue data total amount for calculating current server node i is Bi；
C2, in the case that queue is added without new task in estimation, current server node i completes task time as ti；
C3, traversal requests primary other server nodes, and records the time responded, wherein the shortest response time is denoted as
min-dti；
C4, according to calculating formulaObtain the priority balance parameters C of server node isi；
C5, according to C 'i=Ci-CsiMore amount of new data cost Ci；
C6, with the C ' of step C5iAs data volume cost CiIt brings step M4 into, completes the server node choosing of task data transfer
It selects.
7. a kind of electric power CPS dynamic load distribution method according to claim 4, which is characterized in that step M4 include with
Lower sub-step:
C1, the job queue data total amount for calculating current server node i is Bi；
C2, in the case that queue is added without new task in estimation, current server node i completes task time as ti；
C3, traversal requests primary other server nodes, and records the time responded, wherein the shortest response time is denoted as
min-dti；
C4, according to calculating formulaObtain the priority balance parameters C of server node isi；
C5, according to C 'i=Ci-CsiMore amount of new data cost Ci；
C6, with the C ' of step C5iAs data volume cost CiIt brings step M4 into, completes the server node choosing of task data transfer
It selects.
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 true CN110348681A (en) | 2019-10-18 |
| CN110348681B 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) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110737695A (en) * | 2019-10-08 | 2020-01-31 | 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 |
|---|
| 丁学智: "一种面向分布式服务器集群的动态负载均衡系统的实现", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 姜同全等: "基于动态阈值分配的流媒体边缘云会话迁移策略", 《计算机工程》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110737695A (en) * | 2019-10-08 | 2020-01-31 | 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 |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110348681B (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Saputra et al. | Energy demand prediction with federated learning for electric vehicle networks | |
| CN104969213B (en) | Data flow for low latency data access is split | |
| CN104317650B (en) | A kind of job scheduling method towards Map/Reduce type mass data processing platforms | |
| CN109600798A (en) | Multi-domain resource allocation method and device in a kind of network slice | |
| CN104202254A (en) | An intelligent load balancing method based on a cloud calculation platform server | |
| CN110188872A (en) | A kind of isomery cooperative system and its communication means | |
| CN107093050A (en) | A kind of vehicle dispatching method and system | |
| CN105302858B (en) | A kind of the cross-node enquiring and optimizing method and system of distributed data base system | |
| CN105205052B (en) | A kind of data digging method and device | |
| CN103026575B (en) | For area of computer aided control the multiple energy consumption devices in electric flux network electric flux consume method | |
| CN108076158A (en) | Minimum load route selection method and system based on Naive Bayes Classifier | |
| CN113312177B (en) | Wireless edge computing system and optimizing method based on federal learning | |
| CN110474904A (en) | A kind of Situation Awareness method and system improving prediction | |
| CN108170531A (en) | A kind of cloud data center request stream scheduling method based on depth belief network | |
| CN107423881A (en) | task distribution method and device | |
| CN104283963A (en) | Distributed and collaborative type CDN load balancing method | |
| Khodar et al. | New scheduling approach for virtual machine resources in cloud computing based on genetic algorithm | |
| CN110348681A (en) | A kind of electric power CPS dynamic load distribution method | |
| CN103580922B (en) | A kind of powerline network application server overload controlling method based on hybrid optimization | |
| CN104301241A (en) | SOA dynamic load distribution method and system | |
| CN107995126A (en) | Load balancing route analysis method and load balancing route analyzer | |
| CN112785770B (en) | Dynamic entity queuing model construction method based on time series | |
| CN105262702B (en) | TDMA communication network slot uniform distribution method based on minimal time delay shake | |
| CN103533052A (en) | Online control method for accessing and distributing tasks in group intelligent perception system | |
| Liu et al. | Joint optimization for bandwidth utilization and delay based on particle swarm optimization |
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 |