CN103793281A - Load balancing method of compute-intensive simulation task - Google Patents
Load balancing method of compute-intensive simulation task Download PDFInfo
- Publication number
- CN103793281A CN103793281A CN201410035347.XA CN201410035347A CN103793281A CN 103793281 A CN103793281 A CN 103793281A CN 201410035347 A CN201410035347 A CN 201410035347A CN 103793281 A CN103793281 A CN 103793281A
- Authority
- CN
- China
- Prior art keywords
- federal member
- load balancing
- load
- artificial tasks
- module
- 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.)
- Pending
Links
Images
Abstract
The invention discloses a load balancing method of a compute-intensive simulation task. The method comprises the following steps that first, a simulation task description module inputs the number of simulation instantiation models this time and the parameters of all the models through a human-computer interaction interface, so that a simulation task description file is generated; second, a load balancing control module reads the simulation task description file, task distribution is carried out by combining the current computer resource utilization condition, and a federate member configuration file is generated; third, starting of federate members and loading of the model parameters are achieved by a federate member dispatching module. By means of the method, load balancing of the compute-intensive simulation task is achieved, and the application result shows that a load balancing function is added in a simulation system based on HLA, a large-sized simulation system can operate efficiently and stably, and the application prospect is wide.
Description
Technical field
The present invention relates to a kind of implementation method of artificial tasks load balancing, particularly relate to a kind of load-balancing method of computation-intensive artificial tasks.
Background technology
Along with the distributed emulation based on HLA is realized the increasingly mature of technology, Large Scale Distributed Simulations has become the main flow of current emulation technology development.But in the time of the large artificial system development based on HLA, the loading problem not existing in taking into account system.In the time of simulation modeling, just create federation and federal member according to the physical model of system or mathematical model, this distributes uneven problem with regard to existing serious load.In simulation run process, HLA does not have the mechanism of load balance yet, easy like this cause whole analogue system due to the overload of certain node slow in reacting even paralysis.Load balance problem becomes more and more outstanding in large-scale distributed emulation, has become the key issue that affects simulation efficiency and correctness.In order to address this problem, just need to introduce balancing method of loads, make system can dynamically optimize existing artificial resource, reduce overall emulation cost.
Summary of the invention
For above the deficiencies in the prior art, the object of the present invention is to provide a kind of load-balancing method of computation-intensive artificial tasks, the problem existing with the Large Scale Distributed Simulations operational efficiency solving based on HLA.
Object of the present invention is achieved through the following technical solutions:
A load-balancing method for computation-intensive artificial tasks, the concrete steps of this equalization methods are:
1) complete the input of artificial tasks by man-machine interface, call load balancing module and generate federal member configuration parameter, complete the structure of the distributing emulation system of load-balancing technique by federal member scheduler module,
2) artificial tasks describing module, by the model quantity of this simulation example of input and the parameter of each model, generates artificial tasks description document;
3) load balancing control module reads artificial tasks description document, carries out task distribution in conjunction with current computer resource utilization, generates federal member configuration file, and intensive simulation calculation task is distributed to multiple federal member operation processes.
4) input parameter of federal member scheduler module using federal member configuration file as the process of establishment, starts federal member executive process, completes the loading of this federal member simulation calculation task configuration parameter.
Described distributing emulation system comprises:
Artificial tasks describing module: input by human-computer interaction interface, form artificial tasks description document;
Load balancing control module: according to current computer cpu busy percentage and memory configurations, generate the configuration parameter of federal member, call federal member scheduler module, the intensive federal member example of D ynamic instantiation;
Federal member scheduler module: the federal member configuration parameter generating according to load balancing control module, the startup and the parameter that complete federal member load.
The invention has the advantages that:
This method can realize the load balancing of computation-intensive artificial tasks, its outstanding feature is definition and the distribution of finishing the work by definition task description file and federal member configuration file, in the analogue system based on HLA, add load balance function, large artificial system can efficiently and stably be moved.
Accompanying drawing explanation
The load-balancing method method flow diagram of a kind of computation-intensive artificial tasks of Fig. 1.
1. artificial tasks describing module 2. load balancing control module 3. federal member scheduler modules
Embodiment
If Fig. 1 is a kind of load-balancing method method flow diagram of computation-intensive artificial tasks, the concrete steps of this equalization methods are:
The first step completes the input of artificial tasks by man-machine interface, call load balancing module and generate federal member configuration parameter, the structure that completes the distributing emulation system of load-balancing technique by federal member scheduler module, comprising: artificial tasks describing module, load balancing control module and federal member scheduler module.Wherein:
The function of artificial tasks describing module is: input by human-computer interaction interface, form artificial tasks description document;
The function of load balancing control module is: federal member is as a node of HLA simulation run, load balancing module is according to current computer cpu busy percentage and memory configurations, generate the configuration parameter of federal member, call federal member scheduler module, the intensive federal member example of D ynamic instantiation;
The function of federal member scheduler module is: the federal member configuration parameter generating according to load balancing control module, the startup and the parameter that complete federal member load.
Second step artificial tasks describing module generates artificial tasks description document
Artificial tasks describing module is a personal-machine interactive interface, by inputting the model quantity of this simulation example and the parameter of each model, generates artificial tasks description document, and this file has been described the computation complexity of this artificial tasks.
The 3rd step load balancing control module generates federal member configuration file
Load balancing control module reads artificial tasks description document, in conjunction with current computer resource utilization, adopt the mode of " dealing out the cards " to carry out task distribution, generate federal member configuration file, intensive simulation calculation task is distributed to multiple federal member operation processes, thereby reaches the load balancing of analogue system operation.This cultural element comprises:
Federal member name: federal unique identification, adopts ASCII character to represent;
Federal member Model instantiation number: the task quantity that federal member distributes, adopts integer number to represent;
Federal member model parameter (1 ... N): model configuration parameter.
Startup and parameter that the 4th step federal member scheduler module completes federal member load
The input parameter of federal member scheduler module using federal member configuration file as the process of establishment, starts federal member executive process, completes the loading of this federal member simulation calculation task configuration parameter.
Should be appreciated that the above detailed description of technical scheme of the present invention being carried out by preferred embodiment is illustrative and not restrictive.Those of ordinary skill in the art modifies reading the technical scheme that can record each embodiment on the basis of instructions of the present invention, or part technical characterictic is wherein equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.
Claims (2)
1. a load-balancing method for computation-intensive artificial tasks, is characterized in that, the concrete steps of this equalization methods are:
1) complete the input of artificial tasks by man-machine interface, call load balancing module and generate federal member configuration parameter, complete the structure of the distributing emulation system of load-balancing technique by federal member scheduler module,
2) artificial tasks describing module, by the model quantity of this simulation example of input and the parameter of each model, generates artificial tasks description document;
3) load balancing control module reads artificial tasks description document, carries out task distribution in conjunction with current computer resource utilization, generates federal member configuration file, and intensive simulation calculation task is distributed to multiple federal member operation processes.
4) input parameter of federal member scheduler module using federal member configuration file as the process of establishment, starts federal member executive process, completes the loading of this federal member simulation calculation task configuration parameter.
2. the load-balancing method of a kind of computation-intensive artificial tasks according to claim 1, is characterized in that, comprising: described distributing emulation system comprises:
Artificial tasks describing module: input by human-computer interaction interface, form artificial tasks description document;
Load balancing control module: according to current computer cpu busy percentage and memory configurations, generate the configuration parameter of federal member, call federal member scheduler module, the intensive federal member example of D ynamic instantiation;
Federal member scheduler module: the federal member configuration parameter generating according to load balancing control module, the startup and the parameter that complete federal member load.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410035347.XA CN103793281A (en) | 2014-01-24 | 2014-01-24 | Load balancing method of compute-intensive simulation task |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410035347.XA CN103793281A (en) | 2014-01-24 | 2014-01-24 | Load balancing method of compute-intensive simulation task |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103793281A true CN103793281A (en) | 2014-05-14 |
Family
ID=50668991
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410035347.XA Pending CN103793281A (en) | 2014-01-24 | 2014-01-24 | Load balancing method of compute-intensive simulation task |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103793281A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104821110A (en) * | 2014-12-31 | 2015-08-05 | 国网电力科学研究院武汉南瑞有限责任公司 | Safety warning education simulation training system based on simulation operation and method thereof |
| CN105426247A (en) * | 2015-11-09 | 2016-03-23 | 张博 | HLA federate planning and scheduling method |
| CN105867996A (en) * | 2015-01-22 | 2016-08-17 | 北京仿真中心 | Case based dynamic construction system and method for simulation system |
| CN108153921A (en) * | 2016-12-05 | 2018-06-12 | 北京仿真中心 | A kind of Dynamical Deployment distribution method of HLA federal members |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040044513A1 (en) * | 2002-09-02 | 2004-03-04 | Noriaki Kitahara | Distributed simulation system |
| CN101226484A (en) * | 2007-12-07 | 2008-07-23 | 华中科技大学 | Method for automatically disposing simulation scene based on simulation gridding |
| CN101741906A (en) * | 2009-12-08 | 2010-06-16 | 中国运载火箭技术研究院 | Grid resource management system supporting HLA distribution interactive simulation and implementation method thereof |
| CN102708232A (en) * | 2012-04-24 | 2012-10-03 | 中国人民解放军国防科学技术大学 | Processing method and device for distributed simulation data |
-
2014
- 2014-01-24 CN CN201410035347.XA patent/CN103793281A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040044513A1 (en) * | 2002-09-02 | 2004-03-04 | Noriaki Kitahara | Distributed simulation system |
| CN101226484A (en) * | 2007-12-07 | 2008-07-23 | 华中科技大学 | Method for automatically disposing simulation scene based on simulation gridding |
| CN101741906A (en) * | 2009-12-08 | 2010-06-16 | 中国运载火箭技术研究院 | Grid resource management system supporting HLA distribution interactive simulation and implementation method thereof |
| CN102708232A (en) * | 2012-04-24 | 2012-10-03 | 中国人民解放军国防科学技术大学 | Processing method and device for distributed simulation data |
Non-Patent Citations (1)
| Title |
|---|
| 翁超: "面向LP网络的HLA分布式仿真负载平衡问题研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104821110A (en) * | 2014-12-31 | 2015-08-05 | 国网电力科学研究院武汉南瑞有限责任公司 | Safety warning education simulation training system based on simulation operation and method thereof |
| CN105867996A (en) * | 2015-01-22 | 2016-08-17 | 北京仿真中心 | Case based dynamic construction system and method for simulation system |
| CN105426247A (en) * | 2015-11-09 | 2016-03-23 | 张博 | HLA federate planning and scheduling method |
| CN105426247B (en) * | 2015-11-09 | 2018-11-06 | 张博 | A kind of HLA federal members programming dispatching method |
| CN108153921A (en) * | 2016-12-05 | 2018-06-12 | 北京仿真中心 | A kind of Dynamical Deployment distribution method of HLA federal members |
| CN108153921B (en) * | 2016-12-05 | 2021-06-04 | 北京仿真中心 | Dynamic deployment and distribution method for HLA (high level architecture) federal members |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ahmad et al. | A hybrid genetic algorithm for optimization of scheduling workflow applications in heterogeneous computing systems | |
| CN103078941B (en) | A kind of method for scheduling task of distributed computing system | |
| CN102508639B (en) | Distributed parallel processing method based on satellite remote sensing data characteristics | |
| CN106164867B (en) | Incremental parallel processing of data | |
| CN109993299A (en) | Data training method and device, storage medium, electronic device | |
| CN105117286A (en) | Task scheduling and pipelining executing method in MapReduce | |
| CN104317749B (en) | Information write-in method and device | |
| CN104506620A (en) | Extensible automatic computing service platform and construction method for same | |
| CN103488775A (en) | Computing system and computing method for big data processing | |
| CN103970609A (en) | Cloud data center task scheduling method based on improved ant colony algorithm | |
| CN102012840A (en) | Batch data scheduling method and system | |
| CN111858027B (en) | Cooperative processing method and system for software robot | |
| CN104331321A (en) | Cloud computing task scheduling method based on tabu search and load balancing | |
| CN103793281A (en) | Load balancing method of compute-intensive simulation task | |
| CN105094984A (en) | Resource scheduling method and system | |
| CN107203421A (en) | A kind of adaptive work in combination stream method in cloud computing environment | |
| CN111158800B (en) | Method and device for constructing task DAG based on mapping relation | |
| Alaasam et al. | Scientific micro-workflows: where event-driven approach meets workflows to support digital twins | |
| Zhang et al. | Future manufacturing industry with cloud manufacturing | |
| CN111953788A (en) | Large-scale cloud platform | |
| Li et al. | Smart simulation cloud (simulation cloud 2.0)—the newly development of simulation cloud | |
| Davidrajuh | Solving assembly line balancing problems with emphasis on cost calculations: a petrinets based approach | |
| Liu et al. | BSPCloud: A hybrid distributed-memory and shared-memory programming model | |
| Ma et al. | Cloud-based multidimensional parallel dynamic programming algorithm for a cascade hydropower system | |
| CN104166593A (en) | Method for computing asynchronous and concurrent scheduling of multiple application functions |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140514 |