CN100588197C - Gridding emulation method and its emulator - Google Patents
Gridding emulation method and its emulator Download PDFInfo
- Publication number
- CN100588197C CN100588197C CN200510137114A CN200510137114A CN100588197C CN 100588197 C CN100588197 C CN 100588197C CN 200510137114 A CN200510137114 A CN 200510137114A CN 200510137114 A CN200510137114 A CN 200510137114A CN 100588197 C CN100588197 C CN 100588197C
- Authority
- CN
- China
- Prior art keywords
- entity
- resource
- task
- information
- network
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a network simulation method which comprises: information distribution, network solid construction, analogue program generator, analogue program operation, and analogue resultoutput. The simulator with said invention comprises a information distribution module, a network grid solid construction module, a simulation program generation module, a simulation program operationmodule, and a simulation result output module. The simulator builds models of user solid, mission solid, resource solid, message catalog solid, distributor solid, network solid and statistic analyzessolid, to support accurately simulating the network grid application environment. The simulator uses the support of disperse mission simulated basic structure, via independent process simulates each solid in said network grid system, while the process cycles repeated, it simulates the solid motion and processes the information of external solid. The simulator has better user interface and abundantmode supports, while it can supply relative data statistic and analyze tool to effectively reflect the simulation result.
Description
Technical field
The invention belongs to grid system and analogue technique field thereof, be specifically related to a kind of gridding emulation method and simulator thereof, it is applicable to the research of the simulation evaluation and test of grid architecture, scheduling strategy, application execution performance etc.
Background technology
Along with the continuous development and the renewal of information technology, people are also increasing to the needs of computing capability and data storage capacities, simultaneously, need collaborative work more closely and resource-sharing each other.The core of grid is exactly to realize the shared and effectively polymerization of computational resource, information resources, storage resources and various application, service and DSS in the wide scope.
Designing a grid system is a very complicated system engineering, and it need consider many problems.For example, the validity of resource management and scheduling is the key that influences the grid system success or not in the grid; The resource wide area is shared the isomerism, fail safe and the network performance that bring and also will be can not be ignored; In addition, fault-tolerant ability, extensibility and adaptive ability all are the contents that mesh design person need consider.Therefore, the grid system designer needed before the newly-designed system of actual deployment to guarantee that it has feasibility and desired execution efficient.
Generally speaking, real grid environment is dynamic change, uncontrollable.In such environment, system designer almost can't be investigated all properties problem of system design.Operate to example explanation with grid application, grid application is moved on a virtual machine, this virtual machine is non-existent before operation, and only when using actual motion, grid ability is set up a virtual machine in real time according to the request that the user moves application.Usually, we can't know the concrete condition of this virtual machine, and as possessory relevant information of the topological structure between the node, node type, node etc., and virtual machine also is dynamic change, be embodied in dynamically leaving and adding of resource, and use dynamically enter and withdraw from two aspects.If twice request moved same application program, a back application may repeat original running hardly, and this has also demonstrated fully the uncontrollability and the nonrepeatability of grid system.
The appearance of network simulator has solved above problems.Simulator function is grid running environment of simulation, and this running environment is may command, reusable.We study different problems in the environment of this simulation, such as feasibility and performance issue.In the environment of this simulation, by the configuration different parameters, real simulated goes out the various application scenarioss in the actual environment more, makes the simulation process better authenticity; The more important thing is, same application program can unconfinedly rerun in identical environment, this is performance evaluation basis relatively, and by analyze the result who tests on simulator, the grid researcher can carry out the selection of design and the improvement of system design.
Summary of the invention
The purpose of this invention is to provide a kind of gridding emulation method, the grid system that it can be by the software simulation complicated architectures is for the research of its organizational structure, scheduling strategy, application execution performance etc. provides simulation tool; The present invention also provides the simulator of realizing according to this method.
A kind of gridding emulation method provided by the invention the steps include:
(1) basic configuration information of simulating grid is set, deploy content comprises analog parameter configuration, user definition, task definition, resources definition, message catalog configuration, scheduler configuration, network configuration and statistical analysis configuration;
(2) the various configuration informations that are provided with according to step (1) are constructed corresponding grid entity, and according to the network connection information of each inter-entity in the network configuration, the network of setting up each inter-entity connects, and forms the organizational structure of simulation grid; The grid entity comprises user subject, task entity, resource entity, message catalog entity, scheduler entities, network entity and statistical analysis entity; Wherein,
User subject is configured and initialization according to user definition information; After simulation beginning, the task that user subject will need to carry out is submitted to corresponding scheduler successively, wait task all finish and receive result after finish;
The task entity uses the static class structure to represent; The task entity is carried out needed various base attribute according to task definition information definition task, comprises size of code, the data volume that needs are handled and the data volume of handling the back result that task is carried out, and the executive condition of task on resource defined;
Resource entity is configured and initialization according to resources definition information; And to the relevant information of message catalog registration resource; Receive the request of query resource multidate information, and return the present information of resource; Reception is from the task of scheduler, according to local scheduling strategy execution task, and the task executions result returned to the user; The term of execution of task, upgrade the change in resources situation of oneself to message catalog; Repeat this process, till receiving end;
The message catalog entity is configured and initialization according to the message catalog configuration information; It receives the log-on message from resource, the related data of available resources in the organization system; According to the setting of resource updates mode, the information of new resources more actively or passively, the situation of change of each resource of Dynamic Maintenance; Message catalog is made response to the inquiry from scheduler, according to the qualified resource of corresponding algorithm search, returns list of available resources; Repeat this process, till receiving end;
Scheduler entities is configured and initialization according to the scheduler configuration information; Scheduler entities receives the user asks the back to message catalog query available resources information, obtains list of available resources; To the resource query multidate information, obtain the current information of resource; Carry out resource selection according to dispatching algorithm, the submission task is given selected resource; Repeat this process, till receiving end;
Network entity comprises router, network traffics generator and network link, is configured and initialization according to network configuration information; Router receives the packet on the link, carry out Route Selection after, again packet is sent; Repeat this process, till receiving end; The network traffics generator produces background traffic according to configuring condition on network, till receiving end;
The statistical analysis entity is configured and initialization according to the statistical analysis configuration information; It collects and writes down all kinds of analogue datas, and statistics is added up and exported to response to the statistical requirements of Various types of data in the simulation process to various index amounts; Response is carried out quantitative analysis and output analysis result to the analysis requirement of simulation process to simulated performance; Repeat this process, till receiving end;
(3) according to the configuration information and the network connection information of entity, generate the simulation program code of above-mentioned each entity, at last each simulation entity and relevant network connection simulation code combination thereof are become complete executable program;
(4) on computer system, carry out above-mentioned simulation program, the operation of simulation grid system;
(5) collect analogue data, statistics, analysis and output analog result.
Network simulator according to said method is realized comprises information collocation module, grid entity structure module, simulation program generation module, simulation program Executive Module and analog result output module; Wherein,
Information collocation module is used to obtain analog parameter and various configuration information, and information offered in the grid entity structure module corresponding module, it is made up of analog parameter configuration module, user definition module, task definition module, resources definition module, message catalog configuration module, scheduler configuration module, Network conf iotag. module and statistical analysis configuration module; Wherein,
The user definition module offers user subject constructing module structure with user configuration information, with the structuring user's entity, is used for simulating the user in the grid;
The task definition module offers task entity structure module with task configuration information, with structure task entity, is used for simulating the task in the grid;
The message catalog configuration module offers message catalog entity structure module with the message catalog configuration information, with the tectonic information bibliographical entity, is used for simulating the message catalog in the grid;
Network conf iotag. module offers the network entity constructing module with network configuration information, with the tectonic network entity, is used for simulating the network in the grid;
The scheduler configuration module offers the scheduler entities constructing module with the scheduler configuration information, with the structure scheduler entities, is used for simulating the scheduler in the grid;
The resources definition module offers the resource entity constructing module with resource allocation information, with the structure resource entity, is used for simulating the resource in the grid;
The statistical analysis configuration module offers statistical analysis entity structure module with the statistical analysis configuration information, with structure statistical analysis entity, is used for obtaining the analog information in the network analog process;
Grid entity structure module is made up of user subject constructing module, task entity structure module, resource entity constructing module, message catalog entity structure module, scheduler entities constructing module, network entity constructing module and statistical analysis entity structure module; Wherein,
The user subject constructing module is used for producing user subject, is used to describe user's behavior;
Task entity structure module is used for producing the task entity, is used to describe user task and various attribute thereof;
The resource entity constructing module is used for producing resource entity, is used for the computational resource and the storage capacity thereof of descriptive system;
Message catalog entity structure module is used for producing the message catalog entity, is used to provide the various information with maintenance resources;
The scheduler entities constructing module is used for producing scheduler entities, is used to finish carry out resource allocation when task is submitted to;
The network entity constructing module is used for producing network entity, comprises router, network traffics generator and network link, is used for the transfer of data of the system that finishes;
Statistical analysis entity structure module is used for producing the statistical analysis entity, is used for obtaining relevant analogue data and statistics and analysis in addition in whole simulation process;
The simulation program generation module generates the simulation program code of above-mentioned each entity according to the configuration information and the network connection information of entity, and each simulation entity and relevant network connection simulation code combination thereof are become complete executable program;
The simulation program Executive Module utilizes the support of discrete event simulation architecture, carries out above-mentioned simulation program on computer system, the operation of simulation grid system;
The analog result output module is collected analogue data, statistical analysis analog result and output.
Network simulator of the present invention has following characteristics: the user interface that (1) is good helps user's rapid modeling; (2) Feng Fu simulation element is supported various types of solid modellings; (3) allow the user to participate in modelling, the existing model library of expansion; (4) the analogue data analysis of detailed analogue data statistics and automation intuitively reflects analog result by chart.
Description of drawings
Fig. 1 is the flow chart of gridding emulation method of the present invention;
Fig. 2 is the flow chart of analog subscriber entity of the present invention;
Fig. 3 simulates the flow chart of resource entity for the present invention;
Fig. 4 is the flow chart of analog information bibliographical entity of the present invention;
Fig. 5 is the flow chart of operation simulation device entity of the present invention;
Fig. 6 is the flow chart of analog network router entity of the present invention;
Fig. 7 is the process chart of statistical analysis entity of the present invention;
Fig. 8 is the sequential chart that simulation inter-entity message of the present invention is handled;
Fig. 9 is the module frame figure of network simulator of the present invention;
Figure 10 is the use flow chart of network simulator of the present invention.
Embodiment
The present invention is described in detail below in conjunction with accompanying drawing and example.
As shown in Figure 1, the step of the inventive method is:
The first step: the basic configuration information of simulation grid is set, and deploy content comprises analog parameter configuration, user definition, task definition, resources definition, message catalog configuration, scheduler configuration, network configuration and statistical analysis configuration.The analog parameter configuration comprises the number of times that simulation process repeats.Comprise the selection of user's number, scheduler and message catalog with the product configuration information.Task configuration information comprises the related information between each user's task number, the attribute of task (size of the result data that obtains after size, the task of the data that the size of task, the needs of executing the task are handled is complete), task.Resource allocation information comprises the number of resource, the attribute of resource (storage capacity of the number of processors that the type of resource, resource have, the operational capability of each processor and resource).The message catalog configuration information comprises the situation that is provided with of each service level in the message catalog and their polymerization methods.The scheduler configuration information comprises the dispatching algorithm that scheduler uses.Network configuration information comprises the definition of router and network traffics generator in the system, the appointment that they are connected with the network of each inter-entity, relevant routing algorithm and network traffics produce pattern.The statistical analysis configuration information comprises the appointment of the parser of the way of output of User Defined statistical indicator amount and statistical method, statistic and statistics.
Second step: the various entities in the structure grid system, and the network of setting up between related entities connects the organizational structure of formation grid system.Its concrete grammar is:
According to the various configuration informations that the first step is provided with, construct corresponding grid entity, and according to the network connection information of each inter-entity in the network configuration, the network of setting up each inter-entity connects, thereby forms the organizational structure of simulation grid.The grid entity comprises user subject, task entity, resource entity, message catalog entity, scheduler entities, network entity and statistical analysis entity.
In gridding emulation method of the present invention, user's behavior comprises: select task dispatcher and transmission tasks request, receive the task handling result.In user subject, the user asks to carry out one or more tasks.For the situation of a plurality of tasks, submit execution successively to according to the order association between task.As shown in Figure 2, user subject is configured and initialization according to user definition information.After simulation beginning, the task that user subject will need to carry out is submitted to corresponding scheduler successively, wait task all finish and receive result after finish.
In gridding emulation method of the present invention, the static state portrayal of an application in the definition grid comprises two parts: problem and the scheduler and the algorithm thereof that address this problem use; And the hypothetical simulation process carries out on task level, and problem is defined by the task of one group of association, and scheduler and algorithm thereof have defined the correlated process of resource selection and task scheduling.The task entity uses the static class structure to represent.The task entity is carried out needed various base attribute according to task definition information definition task, comprises size of code, the data volume that needs are handled and the data volume of handling the back result that task is carried out.In addition, also the executive condition (resource type, computing capability, storage capacity etc.) of task on resource defined.Execution association between the task is described by directed acyclic graph, and the user can generate this figure by wscript.exe automatically by the relevant input file of configuration.
In gridding emulation method of the present invention, resource entity is used for the computational resource and the storage capacity thereof of descriptive system.Suppose that the performance of computational resource describes by its computing capability, each computational resource can have one or more computing nodes, and each computing node can have a plurality of processor units.As shown in Figure 3, resource entity is configured and initialization according to resources definition information.After the initialization, each resource entity is carried out following operation: to the relevant information of message catalog registration resource; Receive the request of query resource multidate information, and return the present information of resource; Reception is from the task of scheduler, according to local scheduling strategy execution task, and the task executions result returned to the user; The term of execution of task, upgrade the change in resources situation of oneself to message catalog.This process repeats always, till receiving end.The time of implementation of task on resource entity determined by following factor: the influence that the local scheduling strategy of the number of the floating-point operation that task comprises, resource, the computing capability of node and I/O processing procedure are carried out task.Resource entity also can receive the message from scheduler, removes waiting for or carrying out of task.
For collection and the maintenance process that reflects global information in the true grid system more accurately, in gridding emulation method of the present invention, abstract according to the function of its realization to message catalog is three parts: ground floor is information service, it will be integrated from the information classification of bottom, form the use information of local resource; The second layer is the aggregation information service, and its effect is that the resource information of a plurality of information services is carried out polymerization, forms message catalog, for query requests provides in a big way information source; The 3rd layer is directory service, and it provides inquiry, discovery and location to resource information towards query requests.In addition, directory service can also directly be sent request to information service, inquires about the relevant information of local resource.As shown in Figure 4, the message catalog entity is configured and initialization according to the message catalog configuration information.It receives the log-on message from resource, the related data of available resources in the organization system; According to the setting of resource updates mode, the information of new resources more actively or passively, the situation of change of each resource of Dynamic Maintenance; Message catalog is made response to the inquiry from scheduler, according to the qualified resource of corresponding algorithm search, returns list of available resources.This process repeats always, till receiving end.
Scheduler is located suitable resource and submission task according to scheduling strategy Query Information catalogue dynamically.The scheduling strategy of simulation process can be defined by the user, can realize various dissimilar strategies by scheduler, can comprise in the strategy multiple to task, user and resource constraint, scheduler entities by the Query Information catalogue to real-time gridding load make a response.As shown in Figure 5, scheduler entities is configured and initialization according to the scheduler configuration information.Its core is exactly a dispatching algorithm, and system provides built-in dispatching algorithm for using, also can be by the User Defined dispatching algorithm.Scheduler needs the Query Information catalogue to obtain the information of available resources in the current system, uses these information, and scheduler is dynamically selected resource and carried out task scheduling.Scheduler receives the user asks the back to message catalog query available resources information, obtains list of available resources; To the resource query multidate information, obtain the current information of resource; Carry out resource selection according to dispatching algorithm, the submission task is given selected resource.This process repeats always, till receiving end.
Network entity comprises router, network traffics generator and network link, and different network models can form by different network entity tissues.As shown in Figure 6, network entity is configured and initialization according to network configuration information.Router receives the packet on the link, carry out Route Selection after, again packet is sent.This process repeats always, till receiving end.The network traffics generator produces background traffic according to configuring condition on network, till receiving end.Connection between the related entities realizes by network link.If there are a plurality of data to transmit on network simultaneously, they adopt the method shared link of timesharing system postulation.The transmission time of data is depended on the size of data, the bandwidth and the delay of network.
The statistical analysis entity is that the collection of Various types of data in the simulation process provides support, multiple base state data are recorded in simulation process, and the built-in and user-defined index amount of system exported intuitively with diagrammatic form, provide analysis result by analysis strategy and algorithm.As shown in Figure 7, the statistical analysis entity is configured and initialization according to the statistical analysis configuration information.The statistical analysis entity is collected all kinds of analogue datas, and they are noted with suitable form; Statistics is added up and exported to response to the statistical requirements of Various types of data in the simulation process to various index amounts; Response is carried out quantitative analysis and output analysis result to the analysis requirement of simulation process to simulated performance.This process repeats always, till receiving end.
The 3rd step: according to the configuration information and the network connection information of entity, generate the simulation program code, the corresponding simulation program code that comprises user subject simulation, the simulation of task entity, resource entity simulation, scheduler entities simulation, the simulation of message catalog entity, network entity simulation and the simulation of statistical analysis entity connects the simulation code combination with each module entity and relevant network thereof at last and becomes complete executable program.
The simulation program code is meant the program code that can carry out according to analog parameter configuration module and the generation of various grid entity structure module on the analog basis framework.
The 4th step: go up the above-mentioned simulation program of execution in computer system (comprising sorts of systems such as PC, work station, cluster), the operation of simulation grid system.The execution of simulation program needs the support of discrete event simulation architecture, and it is a processor-oriented discrete event simulation bag.Based on this simulation package, various entities are carried out function separately in the simulation process.Simulation package is according to the thread of simulation program for an independent operating of each entity generation, and the behavior of entity is described in its thread body method.Simulated time point when in addition, the discrete event simulation bag can also provide each entity executive system behavior and simulation process various controls in time.By obtaining these information, can finish requirement to network analog and statistical analysis thereof.
The time program process that each inter-entity message was handled when as shown in Figure 8, the simulation program code was carried out is:
(1) resource is registered to message catalog, and the message catalog entity receives the log-on message from resource, forms Resource TOC;
(2) user submits to scheduler entities according to configuration successively with task requests;
(3) scheduler Query Information catalogue, and obtain the list of available resources that message catalog returns;
(4) resource in the list of available resources is inquired about its present state respectively;
(5) carry out resource allocation and task scheduling according to the multidate information of the resource of returning and the scheduling strategy of scheduler, task assignment is carried out on adequate resources;
(6) after task executes, task executions result will return to user subject;
(7) in the process that task is carried out on the simulation resource, the statistical analysis entity is collected current various analogue datas to resource entity, is used for follow-up statistics and analysis process;
(8) in simulation process, having some new resources can use, and some resource failed, thereby causes resource performance constantly to change, and this just needs the message catalog entity resource information that upgrades in time, reflects the situation of change in resources.
The 5th step: collect the analogue data that step 4 produced, statistics, analysis and output analog result.According to the various index amounts of dry run result and system configuration, add up respectively; And, the analogue data and the statistics of appointment are analyzed according to relevant parser; At last, export The above results as requested.
As shown in Figure 9, the network simulator of realizing according to said method is made up of information collocation module, grid entity structure module, simulation program generation module, simulation program Executive Module and analog result output module.
Information collocation module is used to obtain analog parameter and various configuration information, and information is offered corresponding module in the grid entity structure module.Information collocation module is made up of analog parameter configuration module, user definition module, task definition module, resources definition module, message catalog configuration module, scheduler configuration module, Network conf iotag. module and statistical analysis configuration module; Wherein,
The user definition module offers user subject constructing module structure with user configuration information, with the structuring user's entity, is used for simulating the user in the grid;
The task definition module offers task entity structure module with task configuration information, with structure task entity, is used for simulating the task in the grid;
The message catalog configuration module offers message catalog entity structure module with the message catalog configuration information, with the tectonic information bibliographical entity, is used for simulating the message catalog in the grid;
Network conf iotag. module offers the network entity constructing module with network configuration information, with the tectonic network entity, is used for simulating the network in the grid;
The scheduler configuration module offers the scheduler entities constructing module with the scheduler configuration information, with the structure scheduler entities, is used for simulating the scheduler in the grid;
The resources definition module offers the resource entity constructing module with resource allocation information, with the structure resource entity, is used for simulating the resource in the grid;
The statistical analysis configuration module offers statistical analysis entity structure module with the statistical analysis configuration information, with structure statistical analysis entity, is used for obtaining the analog information in the network analog process.
Grid entity structure module is made up of user subject constructing module, task entity structure module, resource entity constructing module, message catalog entity structure module, scheduler entities constructing module, network entity constructing module and statistical analysis entity structure module; Wherein,
The user subject constructing module is used for producing user subject, is used to describe user's behavior;
Task entity structure module is used for producing the task entity, is used to describe user task and various attribute thereof;
The resource entity constructing module is used for producing resource entity, is used for the computational resource and the storage capacity thereof of descriptive system;
Message catalog entity structure module is used for producing the message catalog entity, is used to provide the various information with maintenance resources;
The scheduler entities constructing module is used for producing scheduler entities, is used to finish carry out resource allocation when task is submitted to;
The network entity constructing module is used for producing network entity, comprises router, network traffics generator and network link, is used for the transfer of data of the system that finishes;
Statistical analysis entity structure module is used for producing the statistical analysis entity, is used for obtaining relevant analogue data and statistics and analysis in addition in whole simulation process.
The simulation program generation module is according to the configuration information and the network connection information of entity, generate the simulation program code, the corresponding simulation program code that comprises user subject simulation, the simulation of task entity, resource entity simulation, scheduler entities simulation, the simulation of message catalog entity, network entity simulation and the simulation of statistical analysis entity connects the simulation code combination with each module entity and relevant network thereof at last and becomes complete executable program.
The simulation program Executive Module utilizes the support of discrete event simulation architecture, carries out above-mentioned simulation program on computer system, the operation of simulation grid system.
The analog result output module is collected analogue data, statistical analysis analog result and output.
Example:
The configuration of network simulator of the present invention and running describe in detail with an example below with reference to Figure 10:
(1) configure user, resource and network condition
At user, resource and network configuration page or leaf, create 3 users and 10 resources, create 2 routers, preceding two users are connected with a router, third party is connected with another router, links to each other between the router, and all resources all link to each other with two routers respectively; Dispose each resource and all be made up of 3 computing nodes, two computing nodes wherein comprise 4 CPU respectively, and another comprises 2 CPU, and the computing capability of each CPU is 377MIPS, and the sharing mode of CPU is that timesharing is shared; The routing algorithm of configuration router is FIFO, and the bandwidth of network link is 10Mb/s.
(2) configure user task
At user's configuration page, dispose the above-mentioned task of having created the user.For each user disposes following task: assignment instructions is carried out length 5000 * 0.8～5000 * 1.2 hundred ten thousand, data 12.5 * 0.8～12.5 * 1.2 kilobytes that need handle during execution, data 30 * 0.8～30 * 1.2 kilobytes after the processing.
(3) configuration information catalogue
At the information service configuration page, create two information service entities and an aggregation information service entities; Configuration level message catalog, promptly preceding four resources and an information service entity associated, all the other six resources and another information service entity associated, two information service entities are related with the aggregation information service entities.
(4) selected task scheduling strategy
In the task scheduling strategy option, select to adopt the available resources wheel to change strategy and dispatch.
(5) configuration analog parameter and statistical analysis parameter
At the analog parameter configuration page, the number of repetition of configuration simulation process is 2; The acquiescence statistic that the configuration statistical parameter provides for system.
(6) generate simulation code according to configuration
Generate page or leaf at code, click " generation " button, generate simulation code.
(7) carry out simulation code
Carry out page or leaf at code, click " execution " button, carry out simulation code.
(8) statistical analysis as a result and output
At the analog result page or leaf, check various index quantitative statistics results in the simulation process and to the correlation analysis conclusion of simulation grid.
The above only is the specific implementation of the best of the present invention, and implementation method of the present invention is not limited thereto, and any change that does not break away from field of the present invention under the spirit of the present invention all should be contained within the scope of the present invention.
Claims (1)
1, a kind of gridding emulation method the steps include:
(1) basic configuration information of simulating grid is set, deploy content comprises analog parameter configuration, user definition, task definition, resources definition, message catalog configuration, scheduler configuration, network configuration and statistical analysis configuration;
(2) the various configuration informations that are provided with according to step (1) are constructed corresponding grid entity, and according to the network connection information of each inter-entity in the network configuration, the network of setting up each inter-entity connects, and forms the organizational structure of simulation grid; The grid entity comprises user subject, task entity, resource entity, message catalog entity, scheduler entities, network entity and statistical analysis entity; Wherein,
User subject is configured and initialization according to user definition information; After simulation beginning, the task that user subject will need to carry out is submitted to corresponding scheduler entities successively, wait task all finish and receive result after finish;
The task entity uses the static class structure to represent; The task entity is carried out needed various base attribute according to task definition information definition task, comprises size of code, the data volume that needs are handled and the data volume of handling the back result that task is carried out, and the executive condition of task on resource defined;
Resource entity is configured and initialization according to resources definition information, and to the relevant information of message catalog registers entities resource; Resource entity receives the request of query resource multidate information, and return the present information of resource, reception is from the task of scheduler entities, according to local scheduling strategy execution task, and the task executions result returned to user subject, the term of execution of task, upgrade the change in resources situation of oneself to the message catalog entity, repeat this process, till receiving end;
The message catalog entity is configured and initialization according to the message catalog configuration information; The message catalog entity receives the log-on message from resource entity, the related data of available resources in the organization system, setting according to the resource updates mode, the information of new resources more actively or passively, the situation of change of each resource of Dynamic Maintenance, the message catalog entity is made response to the inquiry from scheduler entities, according to the qualified resource of corresponding algorithm search, return list of available resources, repeat this process, till receiving end;
Scheduler entities is configured and initialization according to the scheduler configuration information; Scheduler entities receives the user asks the back to message catalog entity query available resources information, obtain list of available resources, to the resource query multidate information, obtain the current information of resource, carry out resource selection according to dispatching algorithm, the submission task is given selected resource, repeats this process, till receiving end;
Network entity comprises router, network traffics generator and network link, is configured and initialization according to network configuration information; Router receives the packet on the link, carry out Route Selection after, again packet is sent, repeat this process, till receiving end; The network traffics generator produces background traffic according to configuring condition on network, till receiving end;
The statistical analysis entity is configured and initialization according to the statistical analysis configuration information; All kinds of analogue datas are collected and write down to the statistical analysis entity, response is to the statistical requirements of Various types of data in the simulation process, statistics is added up and exported to various index amounts, response is to the analysis requirement of simulation process, simulated performance is carried out quantitative analysis and output analysis result, repeat this process, till receiving end;
(3) according to the configuration information and the network connection information of entity, generate the simulation program code of above-mentioned each entity, at last each simulation entity and relevant network connection simulation code combination thereof are become complete executable program;
(4) on computer system, carry out above-mentioned simulation program, the operation of simulation grid system;
(5) collect analogue data, statistics, analysis and output analog result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200510137114A CN100588197C (en) | 2005-12-31 | 2005-12-31 | Gridding emulation method and its emulator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200510137114A CN100588197C (en) | 2005-12-31 | 2005-12-31 | Gridding emulation method and its emulator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1805424A CN1805424A (en) | 2006-07-19 |
| CN100588197C true CN100588197C (en) | 2010-02-03 |
Family
ID=36867262
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200510137114A Expired - Fee Related CN100588197C (en) | 2005-12-31 | 2005-12-31 | Gridding emulation method and its emulator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100588197C (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7493235B2 (en) * | 2007-01-09 | 2009-02-17 | Dell Products, Lp | System and method for dynamic generation of environmental operational models |
| CN100498702C (en) * | 2007-04-29 | 2009-06-10 | 华为技术有限公司 | Configuring information processing method and arrangement, and platform system |
| CN101447897B (en) * | 2007-11-27 | 2011-05-04 | 中国科学院计算技术研究所 | Wireless sensor network simulator and method for establishing the same |
| GB2469509B8 (en) * | 2009-04-16 | 2012-05-30 | Aircom Internat Ltd | Modelling apparatus and method |
| CN101997704A (en) * | 2009-08-20 | 2011-03-30 | 中兴通讯股份有限公司 | Method, device and network management server for realizing element location link state polling |
| CN102346681B (en) * | 2010-07-26 | 2014-11-26 | 中国人民解放军国防科学技术大学 | General-purpose simulator |
| CN104168155B (en) * | 2014-07-31 | 2019-02-15 | 南京赛拜诺信息技术有限公司 | Application traffic generation method and device |
| CN110610089B (en) * | 2019-08-16 | 2023-02-28 | 创新先进技术有限公司 | User behavior simulation method and device and computer equipment |
-
2005
- 2005-12-31 CN CN200510137114A patent/CN100588197C/en not_active Expired - Fee Related
Non-Patent Citations (2)
| Title |
|---|
| 网格资源管理和调度仿真工具研究进展. 李波，石冰心，沈斌，刘继承.微型机与应用，第3期. 2005 |
| 网格资源管理和调度仿真工具研究进展. 李波，石冰心，沈斌，刘继承.微型机与应用，第3期. 2005 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1805424A (en) | 2006-07-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100588197C (en) | Gridding emulation method and its emulator | |
| CN103955373B (en) | A kind of method for designing of SDN application integration development environment | |
| US20050204333A1 (en) | Integrated system-of-systems modeling environment and related methods | |
| Lin et al. | ABS-YARN: A formal framework for modeling Hadoop YARN clusters | |
| Aksakalli et al. | Systematic approach for generation of feasible deployment alternatives for microservices | |
| CN100531070C (en) | Network resource scheduling simulation system | |
| US8443073B2 (en) | Automated performance prediction for service-oriented architectures | |
| Ataie et al. | Modeling and evaluation of dispatching policies in IaaS cloud data centers using SANs | |
| Pagliari et al. | Engineering cyber‐physical systems through performance‐based modelling and analysis: A case study experience report | |
| Pop et al. | Performance analysis of grid DAG scheduling algorithms using MONARC simulation tool | |
| Xu et al. | Modeling the Execution Architecture of a Mobile Phone Software System by Colored Petri Nets | |
| Pop et al. | A simulation model for grid scheduling analysis and optimization | |
| Haeri et al. | VNE-Sim: a virtual network embedding simulator | |
| Hamida et al. | Integrated CHOReOS middleware-Enabling large-scale, QoS-aware adaptive choreographies | |
| Skałkowski et al. | QoS-based storage resources provisioning for grid applications | |
| CN101281553B (en) | Vehicle durability distributed emulation graticule application system | |
| Amar et al. | Tunable scheduling in a GridRPC framework | |
| CN113656952B (en) | Modeling tool for cloud edge computing environment | |
| Li et al. | A direct execution approach to simulating mobile agent algorithms | |
| Waters et al. | Permabase: Predicting the performance of distributed systems at the design stage | |
| Haque et al. | Simulation of a complex distributed real-time database system | |
| CN115941504B (en) | Modeling and simulation method for operation and treatment of micro-service system | |
| Jin et al. | JFreeSim: a grid simulation tool based on MTMSMR model | |
| Feoktistov et al. | Methods and tools for evaluating the reliability of information and computation processes in Grid and cloud systems. | |
| Guirado et al. | Performance prediction using an application-oriented mapping tool |
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 | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100203 Termination date: 20121231 |