CN103020197B - Grid simulation platform and grid simulation method - Google Patents
Grid simulation platform and grid simulation method Download PDFInfo
- Publication number
- CN103020197B CN103020197B CN201210512306.6A CN201210512306A CN103020197B CN 103020197 B CN103020197 B CN 103020197B CN 201210512306 A CN201210512306 A CN 201210512306A CN 103020197 B CN103020197 B CN 103020197B
- Authority
- CN
- China
- Prior art keywords
- grid
- database
- server
- calculating
- web server
- 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
Abstract
The invention discloses a grid simulation platform, which comprises a terminal user, a grid server and computing nodes. After the terminal user logs onto a grid portal through a browser, grid services, such as job submission, information inquiry and the like, can be used. The grid server comprises a Web server, a dispatch server and a database. The grid computing nodes comprise all personal computers (PC) and cluster systems participating in computing. The invention also provides a corresponding grid simulation method. According to the grid simulation platform and the grid simulation method disclosed by the invention, computing of population fitness values is reasonably parallelized, and therefore, resource sharing of distributed desktop systems and cluster systems is realized, the problem that product development and production are influenced due to the fact that a single enterprise resource is insufficient at present is solved, the resource utilization rate is effectively improved, and a large number of funds are saved for enterprises.
Description
Technical field
The present invention relates to computer science and technology field, the management in idle time ability being specifically related to a kind of CPU of the computing machine utilized on internet solves the mesh emulation platform of mass computing problem, and corresponding grid simulation method.
Background technology
In recent years, the superperformance that genetic algorithm has in function optimization makes it demonstrate fabulous application prospect in the structure optimization field of continuously/discrete heat sources.On the other hand, along with the development of gridding technique, people by the computational resource of dispersion by network polymerization, can be formed supernet distribution computing power, concentrate and solve large-scale calculations problem, pellucidly for user provides service.P2P(Peer-to-Peer) pattern, i.e. ad-hoc mode, be a kind of network calculations form be widely adopted, people can utilize the idle processor cycle of all Net-connected computers, and polymerization computing power, completes the task of computation-intensive.
Mesh emulation design incorporates P2P computing thought on the basis of computing grid technology, the computational resource of dispersion in LAN (Local Area Network) even Internet is integrated into virtual high-performance computing environment, set up the genetic algorithm computation model towards multiobjective topology optimization, unified computational resource service and visual user environment for use are provided.Target utilizes grid computing technology to provide a cheapness and effective solution to be used for running optimizatin task on Parallel and Distributed Systems, to help the optimization time shortening the design phase, improve the utilization factor of computational resource, reduce a large amount of human and material resources consumption, reduce manufacturing cost.
Find also there is no ripe mesh emulation platform design at present after deliberation.In view of this, design a kind of mesh emulation that realizes and calculate and really available platform and method, become a kind of demand of reality.
Summary of the invention
The object of the present invention is to provide a kind of mesh emulation platform and grid simulation method, can reasonably by the calculating parallelization of Population adaptation angle value, realize the desktop system of distribution and the resource sharing of group system, alleviate the problem affecting product development and production due to individual enterprise's inadequate resource, and effectively improve resource utilization, for enterprise saves fund.
Technical scheme of the present invention is as follows:
The invention provides a kind of mesh emulation platform, its composition comprises:
One or more terminal user, is connected with the Web server in grid service server respectively, optimizes Job execution parameter for submitting to grid service server, and to grid service server query optimization result;
A grid service server, comprises a database, and respectively with a Web server and a dispatch server of described DataBase combining;
Described database, for preserving the various data messages optimizing operation;
Described Web server, is connected with terminal user and database, and for the optimization Job execution parameter that receiving terminal user submits to, information is optimized in the registration in calling data storehouse, generates list record in a database, and optimization is added job queue;
Described dispatch server, is connected with calculating crunode and database, for extracting the information in database, selecting team's head operation, producing initial population, carry out loop iteration according to the content recorded in list; And individual goal function is calculated encapsulation, distribute to each calculating crunode, and arrange the evaluation result that each calculating crunode returns, generate population of future generation; And optimum results is recorded in database;
Multiple calculating crunode, comprising all PCs and group system that participate in calculating, being connected respectively, for completing the population at individual objective function calculation task being packaged into multiple independently TU task unit with the dispatch server in grid service server.
The present invention also provides a kind of grid simulation method, comprises the following steps:
Step 1: user submits initialization function by browser to Web server;
Step 2:Web server calls database interface registration optimization information, generates list record in a database by Web server, and this suboptimization is added job queue;
Step 3: dispatch server extracts database information, selects team's head operation, produces initial population, enter loop iteration according to the content recorded in list;
Step 4: the objective function of population at individual calculates and is packaged into multiple independently TU task unit by dispatch server, distributes to calculating crunode, have been walked abreast calculating by calculating crunode all in grid;
Step 5: dispatch server arranges after collecting the evaluation result that whole calculating crunode returns, and generates population of future generation and evaluates it;
Step 6: iterative process is until find optimum solution or reach end condition and just terminate, and dispatch server by outcome record in a database;
Step 7, user is to Web server query optimization result.
Advantageous Effects of the present invention is:
One, by the present invention, can improve or the efficiency of all computational resources and utilization factor in expanding type enterprise, meet the demand of final user, before can solving simultaneously owing to calculating, the shortage of data or storage resources and insurmountable problem.
Two, by the present invention, can Virtual Organization be set up, by sharing application and data, common problem be cooperated.
Three, by the present invention, can conformity calculation ability, storage and other resources, the huge problem solving needing a large amount of computational resource can be made to become possibility.
Four, by the present invention, can share resource, effectively optimize and holistic management, the total cost of calculating can be reduced.
Accompanying drawing explanation
Fig. 1 is the system assumption diagram of mesh emulation platform of the present invention.
Fig. 2 is the hardware configuration schematic diagram of mesh emulation platform of the present invention.
Embodiment
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described further.
As shown in Figure 1, mesh emulation platform adopts three-layer architecture: comprise grid application layer, grid services layer and gridding resource layer.Grid application layer utilizes grid portal (Grid Portal) to provide the Web interface of access grid resource to user, and terminal user can use the mesh services such as Hand up homework, information inquiry by browser; Grid services layer, primarily of Web server and dispatch server composition, realizes information service and job service by database service interface and application program; Gridding resource layer comprises all computational resources in grid environment, carries out communication and data transmission between level based on http agreement.
Fig. 2 shows structure deployment diagram of the present invention.As shown in Figure 2, hardware composition of the present invention is mainly divided into three parts: terminal user, grid service server and calculating crunode.
Terminal user can be one, also can be multiple, and each terminal user is connected with the Web server in grid service server respectively.After terminal user signs in grid portal by browser, can to Web server submission optimization operation and to Web server query optimization object information.
Grid service server comprises Web server, dispatch server and database.Database preserves the various data messages optimizing operation.Web server and dispatch server realize information service and job service by database service interface and application program.Wherein, Web server is connected with terminal user and database, the optimization Job execution parameter that its receiving terminal user submits to, and information is optimized in the registration in calling data storehouse, generates list record in a database, and optimization is added job queue.Dispatch server is connected with calculating crunode and database, and it extracts the information in database, selects team's head operation, produces initial population, carry out loop iteration according to the content recorded in list; And individual goal function is calculated encapsulation, distribute to each calculating crunode, and arrange the evaluation result that each calculating crunode returns, generate population of future generation; And optimum results is recorded in database.
Calculating crunode has multiple, comprises in grid environment all PCs and group system that participate in calculating.Each calculating crunode is connected with dispatch server respectively, completes the population at individual objective function calculation task being packaged into multiple independently TU task unit.
Below in conjunction with Fig. 2, set forth the workflow methods of mesh emulation platform job service of the present invention.
As denoted by the arrows in fig. 2, first in step 1, user logs in grid portal by browser and submits initialization function to Web server, comprises the various Job execution parameters such as Population Size.Then in step 2, Web server calling data bank interface registration optimization information, generates the list record of specific format in a database by Web server, and this suboptimization is added job queue.Next in step 3, dispatch server extracts database information, selects team's head operation, produces initial population according to the content recorded in list, starts first time iteration.In step 4, the objective function of population at individual calculates and is packaged into multiple independently TU task unit by dispatch server, distributes to calculating crunode, have been walked abreast calculating by calculating crunode all in grid.In steps of 5, dispatch server arranges after collecting the evaluation result that whole calculating crunode returns, and generates population of future generation and evaluates it.In step 6, iterative process is until find optimum solution or reach end condition and just terminate, and dispatch server by outcome record in a database; In step 7, user inquires about final optimum results by grid portal to Web server finally.
In sum, mesh emulation platform of the present invention can realize following major function:
One, user management: registration, login, authentication.
Two, Hand up homework: user uploads initialization files, carries out optimum configurations, and after confirmation, this operation adds queuing system wait scheduling automatically.
Three, job state: the implementation progress of user job and the details (transmitting time, time of return, computation host, state etc.) of each TU task unit can be checked.
Four, result queries: user can check that job output file comprises the intermediate result file of optimization, the automatic formation curve figure of optimum results.
Five, performance monitoring: whole gridding resource Static and dynamic information can be shown, comprise resource distribution, machine loading, network performance etc.
Above-described is only the preferred embodiment of the present invention, the invention is not restricted to above embodiment.Be appreciated that the oher improvements and changes that those skilled in the art directly derive without departing from the basic idea of the present invention or associate, all should think and be included within protection scope of the present invention.
Claims (2)
1. a mesh emulation platform, is characterized in that comprising:
One or more terminal user, is connected with the Web server in grid service server respectively, optimizes Job execution parameter for submitting to grid service server, and to grid service server query optimization result;
A grid service server, comprises a database, and respectively with a Web server and a dispatch server of described DataBase combining;
Described database, for preserving the various data messages optimizing operation;
Described Web server, is connected with terminal user and database, and for the optimization Job execution parameter that receiving terminal user submits to, information is optimized in the registration in calling data storehouse, generates list record in a database, and optimization is added job queue;
Described dispatch server, is connected with calculating crunode and database, for extracting the information in database, selecting team's head operation, producing initial population, carry out loop iteration according to the content recorded in list; And individual goal function is calculated encapsulation, distribute to each calculating crunode, and arrange the evaluation result that each calculating crunode returns, generate population of future generation; And optimum results is recorded in database;
Multiple calculating crunode, comprise all PCs and group system that participate in calculating, be connected with the dispatch server in grid service server respectively, the objective function of population at individual calculates and is packaged into multiple independently TU task unit by dispatch server, distribute to calculating crunode, to have been walked abreast calculating by calculating crunode all in grid.
2., based on the grid simulation method of mesh emulation platform described in claim 1, it is characterized in that comprising the following steps:
Step 1: user submits initialization function by browser to Web server;
Step 2:Web server calls database interface registration optimization information, generates list record in a database by Web server, and this suboptimization is added job queue;
Step 3: dispatch server extracts database information, selects team's head operation, produces initial population, enter loop iteration according to the content recorded in list;
Step 4: the objective function of population at individual calculates and is packaged into multiple independently TU task unit by dispatch server, distributes to calculating crunode, have been walked abreast calculating by calculating crunode all in grid;
Step 5: dispatch server arranges after collecting the evaluation result that whole calculating crunode returns, and generates population of future generation and evaluates it;
Step 6: iterative process is until find optimum solution or reach end condition and just terminate, and dispatch server by outcome record in a database;
Step 7, user is to Web server query optimization result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210512306.6A CN103020197B (en) | 2012-12-04 | 2012-12-04 | Grid simulation platform and grid simulation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210512306.6A CN103020197B (en) | 2012-12-04 | 2012-12-04 | Grid simulation platform and grid simulation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103020197A CN103020197A (en) | 2013-04-03 |
| CN103020197B true CN103020197B (en) | 2015-07-08 |
Family
ID=47968801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210512306.6A Active CN103020197B (en) | 2012-12-04 | 2012-12-04 | Grid simulation platform and grid simulation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103020197B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103888537A (en) * | 2014-03-27 | 2014-06-25 | 浪潮电子信息产业股份有限公司 | Method and system for grid computing based on web page |
| WO2024045090A1 (en) * | 2022-08-31 | 2024-03-07 | 西门子股份公司 | Product model simulation method and device |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102724064A (en) * | 2012-05-17 | 2012-10-10 | 清华大学 | Method for building network application simulation system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6944620B2 (en) * | 2002-11-04 | 2005-09-13 | Wind River Systems, Inc. | File system creator |
-
2012
- 2012-12-04 CN CN201210512306.6A patent/CN103020197B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102724064A (en) * | 2012-05-17 | 2012-10-10 | 清华大学 | Method for building network application simulation system |
Non-Patent Citations (1)
| Title |
|---|
| 基于志愿者计算的高性能分布式作业系统及其优化;肖颖等;《计算机工程与设计》;20101231;第3158-3161页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103020197A (en) | 2013-04-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xu et al. | Intelligent resource management in blockchain-based cloud datacenters | |
| He et al. | AMTS: Adaptive multi-objective task scheduling strategy in cloud computing | |
| Wang et al. | Load balancing task scheduling based on genetic algorithm in cloud computing | |
| Zhu et al. | Study on cloud computing resource scheduling strategy based on the ant colony optimization algorithm | |
| Chunlin et al. | Cost and energy aware service provisioning for mobile client in cloud computing environment | |
| CN102404412A (en) | Energy saving method and system for cloud compute data center | |
| Tripathi et al. | Modified dragonfly algorithm for optimal virtual machine placement in cloud computing | |
| Wu et al. | Meccas: Collaborative storage algorithm based on alternating direction method of multipliers on mobile edge cloud | |
| CN104821906A (en) | Efficient energy-saving virtual network node mapping model and algorithm | |
| Zhao et al. | An energy and carbon-aware algorithm for renewable energy usage maximization in distributed cloud data centers | |
| CN103679564A (en) | Task allocation method applicable to power distribution network topology analysis distributed computation | |
| Xu et al. | Fog-cloud task scheduling of energy consumption optimisation with deadline consideration | |
| He et al. | Energy-efficient framework for virtual machine consolidation in cloud data centers | |
| CN103020197B (en) | Grid simulation platform and grid simulation method | |
| Huang et al. | Computation offloading for multimedia workflows with deadline constraints in cloudlet-based mobile cloud | |
| Chen et al. | Research on workflow scheduling algorithms in the cloud | |
| CN106610866A (en) | Service value constrained task scheduling algorithm in cloud storage environment | |
| Cao et al. | A resource allocation strategy in fog-cloud computing towards the Internet of Things in the 5g era | |
| CN104702690A (en) | Distributed high-performance computing method based on virtual tree network technology | |
| Liu et al. | Joint optimization for bandwidth utilization and delay based on particle swarm optimization | |
| Bhagavathi et al. | Improved beetle swarm optimization algorithm for energy efficient virtual machine consolidation on cloud environment | |
| CN103179167A (en) | Method and system for cloud computing and load balancing server | |
| Sinha et al. | AI-Driven Task Scheduling Strategy with Blockchain Integration for Edge Computing | |
| Li et al. | Cluster load based content distribution and speculative execution for geographically distributed cloud environment | |
| Li et al. | Edge computing offloading strategy based on dynamic non-cooperative games in D-IoT |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |