CN117390841A - Different-place collaborative simulation platform architecture based on super computing cloud and design method - Google Patents
Different-place collaborative simulation platform architecture based on super computing cloud and design method Download PDFInfo
- Publication number
- CN117390841A CN117390841A CN202311298490.3A CN202311298490A CN117390841A CN 117390841 A CN117390841 A CN 117390841A CN 202311298490 A CN202311298490 A CN 202311298490A CN 117390841 A CN117390841 A CN 117390841A
- Authority
- CN
- China
- Prior art keywords
- simulation
- task
- server
- subtasks
- job
- 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
- 238000004088 simulation Methods 0.000 title claims abstract description 212
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000013461 design Methods 0.000 title claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 22
- 238000012546 transfer Methods 0.000 claims abstract description 9
- 230000008569 process Effects 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000005516 engineering process Methods 0.000 claims description 9
- 238000012163 sequencing technique Methods 0.000 claims description 8
- 238000009434 installation Methods 0.000 claims description 5
- 230000003993 interaction Effects 0.000 abstract description 23
- 230000005540 biological transmission Effects 0.000 description 9
- 230000008901 benefit Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 230000007246 mechanism Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008520 organization Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
A remote collaborative simulation platform architecture and a design method based on a super computing cloud relate to the technical field of simulation platform architecture design. The method solves the problem of data interaction and collaborative operation in complex simulation business. The method comprises the following steps: the task management authority client creates a collaborative simulation task locally, splits the collaborative simulation task into different simulation subtasks and sends the different simulation subtasks to the execution client; the execution client selects simulation software of the super computing cloud according to the simulation subtask requirements and submits the simulation software to a server program deployed on a remote application server; the server program converts the simulation subtasks into computer executable jobs and transmits the jobs to the super computing resource access server in a unified way; the super computing resource access server receives the job to perform resource scheduling processing and sends the job to the data synchronization server; the data synchronization server synchronously transfers the data to the data sharing server, and the data sharing server performs classified storage management. The method is applied to the field of aeroengine simulation.
Description
Technical Field
The invention relates to the technical field of simulation platform architecture design, in particular to a remote collaborative simulation platform architecture design method based on a super computing cloud.
Background
In a large-scale complex simulation service scene in the field of aeroengines, the problems of data interaction, collaborative operation and collaborative operation time sequence control among different simulation software under the collaborative simulation condition are related to the cross-component, multi-disciplinary and cross-region collaborative simulation conditions, so that great challenges are brought to complex simulation services of complex cross-component and multi-disciplinary in the field of aeroengine simulation, the problem of data interaction in the cross-region collaborative simulation service scene is always a pain point problem in the industry, and particularly under the condition of large simulation data, the data cross-region transmission collaboration has no better solution.
Aiming at the co-simulation problem, the existing distributed co-simulation technology can solve the co-simulation service scene with lower complexity to a certain extent, especially the service scene with smaller data interaction quantity, but the method is dependent on a local area network environment, and the problem of cross-region co-operation is not thoroughly solved yet, so that the method is obviously not suitable for simulation application in the field of aeroengines.
How to realize the data interaction, cooperative operation and the like in the complex simulation business is a big bottleneck problem in the field of the prior aero-engine, and is a difficult problem which needs to be solved at present.
Disclosure of Invention
Aiming at solving the problems of data interaction, collaborative operation and the like in complex simulation business, the invention provides a remote collaborative simulation platform architecture based on a super computing cloud, wherein the framework comprises:
the system comprises a client, a server, a super computing resource access server and a data sharing server;
the client comprises a task management authority client and an execution client, and is used for issuing a simulation task and transmitting the simulation task to the server;
the server is used for receiving the simulation task and uploading the simulation task to the super computing resource access server according to the task processing executable by the computer;
and the super computing resource access server receives the job to carry out resource scheduling and stores the resource scheduling data to a data sharing server.
Based on the same inventive concept, the invention also provides a different-place collaborative simulation platform architecture design method based on the super computing cloud, which comprises the following steps:
s1: the task management authority client creates a collaborative simulation task locally, splits the collaborative simulation task into different simulation subtasks, and sends the different simulation subtasks to the execution client according to the parts and disciplines of the simulation subtasks;
s2: the execution client receives the simulation subtask, selects simulation software of the super computing cloud according to the requirement of the simulation subtask, and submits the simulation subtask to a server program deployed on a remote application server through the execution client;
s3: the method comprises the steps that a server program receives a simulation subtask submitted by an execution client, converts the simulation subtask into a job executable by a computer, and uniformly transmits the job to a super computing resource access server;
s4: the super computing resource access server receives the job, performs resource scheduling processing according to the job, and sends simulation data of the scheduling processing to the data synchronization server;
s5: and the data synchronization server receives the simulation data and synchronously transfers the simulation data to the data sharing server, and the data sharing server performs classified storage management according to the data type.
Further, there is also provided a preferred mode, wherein the step S1 includes: when the subtasks are split, subtask numbers are generated for the subtasks according to the subtask execution sequence and the natural number sequence and are used as the basis of the subtask execution sequence.
Further, a preferred manner is also provided, wherein the limited client is developed by adopting QT framework technology.
Further, there is also provided a preferred manner, wherein the converting the simulation subtask into a computer-executable job in the step S3 includes: and storing the simulation task into a job template in a group+key-value mode according to a predefined semantic grammar rule, wherein group is a parameter group, key is a key, and value is a value corresponding to the key.
Further, there is also provided a preferred mode, wherein the predefined semantic grammar rule includes:
the first parameter group is task information, and the key-value pair comprises a total task number, the number of subtasks under the total task and the current subtask number;
the second parameter group is simulation software information, and the key-value pair comprises a simulation software name, a version number, an installation address, driving script information and input file information;
the other parameter sets are simulation software parameter setting information.
Further, there is also provided a preferred mode, wherein the step S4 includes:
after receiving the job request sent by the server program, the super computing resource access service sorts the jobs by taking the total task as a unit, and sends the jobs to the super computing job scheduling system for execution according to the job sequence;
the super-computing operation scheduling system also monitors the execution state of the task in real time in the task execution process, feeds back the execution state and log information to the server program for real-time display, and sends all simulation data path information under the task to the data synchronization server after all sub-tasks are executed.
Further, there is also provided a preferred mode, wherein the sending the job sequence to the super computing job scheduling system for execution includes:
analyzing the operation file;
and reading a task information group, numbering the task information group according to the total task, acquiring the job documents of all the subtasks under the total task, sequencing the subtasks according to the job document numbering sequence of the subtasks, and uniformly submitting the sequencing result to a super-computing job scheduling system for execution.
Based on the same inventive concept, the invention also provides a computer readable storage medium for storing a computer program, wherein the computer program executes the above-mentioned method for designing the remote collaborative simulation platform architecture based on the super computing cloud.
Based on the same inventive concept, the invention also provides a computer device, which comprises a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the above-mentioned method for designing the remote co-simulation platform architecture based on the super computing cloud.
The invention has the advantages that:
the invention solves the problems of data interaction, collaborative operation and the like in complex simulation business.
According to the remote co-simulation platform architecture based on the super computing cloud, data interaction is achieved through the client, the server and the data sharing server; the client comprises a task management authority client and an execution client, wherein the task management authority client is used for issuing a simulation task and transmitting the simulation task to the server, the server receives the simulation task and processes the simulation task into a job executable by a computer, and then uploads the job executable by the computer to the super computing resource access server, and the data sharing server is used for storing data of resource scheduling. By designing appropriate data formats and protocols, and using high-speed network transmission techniques, efficient data transmission and sharing between different components can be performed. The simulation platform architecture provided by the invention realizes cooperative operation by using the super computing resource access server and the task management authority client. The task management authority client of the client is responsible for issuing simulation tasks, and the super computing resource access server is responsible for receiving and processing the tasks. The super computing resource access server performs resource scheduling and distributes the jobs to available computing resources for processing. And through a unified scheduling and control mechanism, the running sequence, parameter transfer and result sharing of different simulation tasks are coordinated, so that collaborative operation is realized. The simulation platform architecture also uses the super computing cloud technology to deploy the super computing resource access server on the cloud, so that remote coordination can be realized. The client, the server and the super computing resource access server can be located in different regions and communicate through the Internet. Super computing clouds provide high performance computing and storage resources and have flexible deployment and expansion capabilities. Through reasonable network design and transmission optimization, the high efficiency and reliability of cross-region data interaction can be ensured.
According to the remote co-simulation platform architecture design method based on the super computing cloud, the co-simulation tasks of cross-components, multidisciplinary and multidimension are realized through the cooperation of the task management authority client and the execution client. The task management authority client can split the collaborative simulation task into different simulation subtasks and send the different simulation subtasks to the execution client, and the execution client selects simulation software of the super computing cloud for processing according to task requirements. The collaborative capability enables complex simulation tasks to be distributed on different execution clients and super computing resource access servers for parallel computation, and simulation efficiency and effect are improved. Further, the super computing resource access server performs resource scheduling and processing according to the requirements of the job after receiving the job. And different simulation subtasks are distributed to available computing resources through a reasonable resource scheduling algorithm, so that optimal utilization of the resources is realized. Therefore, the simulation subtasks can be guaranteed to be fully supported by computing resources in the super computing cloud, and the simulation speed and quality are improved. The invention also uses the data synchronization server and the data sharing server to realize the interaction and storage of the simulation data. After the super computing resource access server processes the simulation subtasks, simulation data are sent to the data synchronization server, and then the data synchronization server transfers the data to the data sharing server. The data sharing server stores the data according to the type in a classified mode, and subsequent data analysis and sharing are facilitated. The data interaction and sharing mechanism ensures data sharing and result transmission among different tasks, and improves the efficiency and accuracy of collaborative simulation.
The method is applied to the field of aeroengine simulation.
Drawings
FIG. 1 is a schematic diagram of a remote co-simulation platform architecture based on a super computing cloud according to an embodiment;
FIG. 2 is a flowchart of a method for designing a remote co-simulation platform architecture based on a super computing cloud according to the second embodiment;
fig. 3 is a flowchart of a super computing resource access service execution according to a seventh embodiment.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments.
Embodiment one, this embodiment will be described with reference to fig. 1. The embodiment of the present invention provides a remote co-simulation platform architecture based on a super computing cloud, where the framework includes:
the system comprises a client, a server, a super computing resource access server and a data sharing server;
the client comprises a task management authority client and an execution client, and is used for issuing a simulation task and transmitting the simulation task to the server;
the server is used for receiving the simulation task and uploading the simulation task to the super computing resource access server according to the task processing executable by the computer;
and the super computing resource access server receives the job to carry out resource scheduling and stores the resource scheduling data to a data sharing server.
In this embodiment, data interaction is implemented through a client, a server, and a data sharing server. The client comprises a task management authority client and an execution client, wherein the task management authority client is used for issuing simulation tasks and transmitting the simulation tasks to the server. The server receives the simulation task and processes the simulation task into a job executable by the computer, and then uploads the job to the super computing resource access server. The data sharing server is used for storing the data of the resource scheduling. By designing appropriate data formats and protocols, and using high-speed network transmission techniques, efficient data transmission and sharing between different components can be performed. The embodiment realizes cooperative operation by using the super computing resource access server and the task management authority client. The task management authority client of the client is responsible for issuing simulation tasks, and the super computing resource access server is responsible for receiving and processing the tasks. The super computing resource access server performs resource scheduling and distributes the jobs to available computing resources for processing. And through a unified scheduling and control mechanism, the running sequence, parameter transfer and result sharing of different simulation tasks are coordinated, so that collaborative operation is realized. In the embodiment, the super computing cloud technology is further used, and the super computing resource access server is deployed on the cloud end, so that remote coordination can be realized. The client, the server and the super computing resource access server can be located in different regions and communicate through the Internet. Super computing clouds provide high performance computing and storage resources and have flexible deployment and expansion capabilities. Through reasonable network design and transmission optimization, the high efficiency and reliability of cross-region data interaction can be ensured.
According to the remote collaborative simulation platform architecture based on the super computing cloud, the problems of data interaction, collaborative operation and the like in complex simulation service are effectively solved through collaborative work of the client, the server, the super computing resource access server and the data sharing server. The method reasonably organizes and distributes task management, resource scheduling and data sharing functions, realizes efficient data interaction and collaborative operation, and is suitable for complex multidisciplinary and cross-region simulation application scenes in the field of aeroengine simulation and the like.
Embodiment two, this embodiment will be described with reference to fig. 2. The embodiment of the method for designing the remote co-simulation platform architecture based on the super computing cloud comprises the following steps:
s1: the task management authority client creates a cross-component, multidisciplinary and multidimensional collaborative simulation task locally, splits the collaborative simulation task into different simulation subtasks, and sends the different simulation subtasks to the execution client according to the components and disciplines of the simulation subtasks;
s2: the execution client receives the simulation subtask, selects simulation software of the super computing cloud according to the requirement of the simulation subtask, and submits the simulation subtask to a server program deployed on a remote application server through the execution client;
s3: the method comprises the steps that a server program receives a simulation subtask submitted by an execution client, converts the simulation subtask into a job executable by a computer, and uniformly transmits the job to a super computing resource access server;
s4: the super computing resource access server receives the job, performs resource scheduling processing according to the job, and sends simulation data of the scheduling processing to the data synchronization server;
s5: and the data synchronization server receives the simulation data and synchronously transfers the simulation data to the data sharing server, and the data sharing server performs classified storage management according to the data type.
According to the embodiment, through cooperation of the task management authority client and the execution client, the cross-component, multidisciplinary and multidimensional collaborative simulation task is realized. The task management authority client can split the collaborative simulation task into different simulation subtasks and send the different simulation subtasks to the execution client, and the execution client selects simulation software of the super computing cloud for processing according to task requirements. The collaborative capability enables complex simulation tasks to be distributed on different execution clients and super computing resource access servers for parallel computation, and simulation efficiency and effect are improved. Further, the super computing resource access server performs resource scheduling and processing according to the requirements of the job after receiving the job. And different simulation subtasks are distributed to available computing resources through a reasonable resource scheduling algorithm, so that optimal utilization of the resources is realized. Therefore, the simulation subtasks can be guaranteed to be fully supported by computing resources in the super computing cloud, and the simulation speed and quality are improved. In the embodiment, a data synchronization server and a data sharing server are also used, so that the interaction and storage of simulation data are realized. After the super computing resource access server processes the simulation subtasks, simulation data are sent to the data synchronization server, and then the data synchronization server transfers the data to the data sharing server. The data sharing server stores the data according to the type in a classified mode, and subsequent data analysis and sharing are facilitated. The data interaction and sharing mechanism ensures data sharing and result transmission among different tasks, and improves the efficiency and accuracy of collaborative simulation.
According to the remote collaborative simulation platform architecture based on the super computing cloud, a client, a server and a super computing resource access server can be deployed in different regions and communicated through the Internet. The remote coordination capability enables team members to perform coordination work in different places, reduces space-time limitation, and improves work efficiency and coordination capability.
The implementation mode realizes the remote co-simulation platform architecture based on the super computing cloud, wherein the remote co-simulation platform architecture comprises functions of task management, resource scheduling, data interaction, sharing and the like. Through the effective collaborative simulation capability, resource scheduling and allocation, data interaction and sharing and the remote collaborative capability, the efficiency and accuracy of the complex simulation service are improved, and the cost is reduced.
In a third embodiment, the present embodiment is further defined by a method for designing a remote co-simulation platform architecture based on a super computing cloud according to the second embodiment, where the step S1 includes: when the subtasks are split, subtask numbers are generated for the subtasks according to the subtask execution sequence and the natural number sequence and are used as the basis of the subtask execution sequence.
In the present embodiment, the execution order of the subtasks can be established by generating subtask numbers for the subtasks and sorting the subtasks according to a natural number sequence. Therefore, the collaborative simulation platform can be ensured to be gradually carried out according to a preset sequence when executing tasks, and confusion and mess are avoided. The sequential execution of the tasks can improve the efficiency and accuracy of task execution, and ensure that the tasks are correctly processed on different execution clients and super computing resource access servers.
According to the method and the device, the subtask numbers are generated for the subtasks, so that the task management and control of the collaborative simulation become more convenient and reliable. The task management authority client can track and monitor subtasks according to subtask numbers, and ensure that the sequence and progress of each subtask accord with expectations. Therefore, the manageability of the task can be improved, the task scheduling and control are facilitated, and the task execution condition is monitored and fed back.
In the embodiment, by generating the subtask numbers for the subtasks, the tasks can be sequentially distributed to different execution clients and the super computing resource access server for processing. The distributed computing support is beneficial to realizing the parallel computing and acceleration of tasks, and improves the simulation efficiency and the running speed. Meanwhile, as the task numbers provide the basis of the execution sequence, each execution client can schedule and process the tasks according to the task numbers, so that the correct execution and coordination of the tasks are ensured.
According to the method and the device, the subtask numbers are generated according to the natural number sequence when the subtasks are split, so that the execution sequence of the tasks can be established, the management and control of the tasks are improved, distributed computing is supported, and the efficiency and accuracy of collaborative simulation are improved.
In the fourth embodiment, the present embodiment is further defined by the method for designing a remote co-simulation platform architecture based on a super computing cloud in the second embodiment, where the client is developed by adopting QT framework technology.
In the embodiment, client software is developed by adopting a QT frame technology, and a collaborative simulation task is created according to service requirements through a visual man-machine interaction interface. The QT framework technology has a cross-platform characteristic, and an application program developed by the QT framework technology can run under a plurality of operating systems such as windows, linux and the like, so that the requirements of different operating systems of different simulation engineers PC machines are met.
In a fifth embodiment, the present embodiment is further defined by a method for designing a remote co-simulation platform architecture based on a super computing cloud according to the second embodiment, where in step S3, the simulation subtasks are converted into computer-executable jobs, including: and storing the simulation task into a job template in a group+key-value mode according to a predefined semantic grammar rule, wherein group is a parameter group, key is a key, and value is a value corresponding to the key.
According to the embodiment, the simulation tasks are organized according to the parameter groups, so that the execution of the tasks can be better managed and controlled. The key value pairs of each parameter set represent specific parameter settings, and parameters of the simulation task can be flexibly adjusted and configured to meet different requirements and scenes. The mode of the group+key-value enables the parameter organization structure to be clear, and is convenient to manage and modify.
According to the method, the defined parameter combination can be reused conveniently by saving the simulation task as the operation template, so that the time and the workload for repeatedly setting the parameters are saved. When similar simulation tasks need to be executed, the saved job templates can be directly used, the process of configuring parameters is reduced, and the working efficiency and consistency are improved.
The predefined semantic grammar rules of the embodiment can ensure the correctness and consistency of parameters. By following the rule, the input range of the parameters can be effectively limited, and illegal parameter combination is prevented. Thus, the reliability and stability of the task can be improved, and errors and anomalies are reduced.
The present embodiment converts the simulation subtasks into computer-executable jobs, meaning that the tasks can be executed directly on the computer without requiring manual individual operations. Therefore, the automation degree and the execution efficiency of the task can be improved, the requirement of manual intervention is reduced, and potential errors caused by human factors are reduced.
The method of converting the simulation subtask into the computer executable job and storing the simulation subtask as the job template by adopting the group+key-value mode can provide the advantages of convenient parameter organization and management, job template reuse, application of semantic grammar rules and computer execution. Such design choices aim to increase the efficiency, accuracy and reliability of task execution and reduce the potential risk of manual operation.
An embodiment six is a further limitation of the method for designing a remote co-simulation platform architecture based on a super computing cloud according to the embodiment five, where the predefined semantic grammar rule includes:
the first parameter group is task information, and the key-value pair comprises a total task number, the number of subtasks under the total task and the current subtask number;
the second parameter group is simulation software information, and the key-value pair comprises a simulation software name, a version number, an installation address, driving script information and input file information;
the other parameter sets are simulation software parameter setting information.
The present embodiment may provide information about the total number of tasks, the number of subtasks, and the current subtask number through the key value pair of the first parameter set. Thus, the execution condition of the tasks can be better managed and tracked, and each subtask can be accurately executed and monitored. The key value pairs in the second parameter set include basic information of emulation software such as name, version number, installation address, driving script information, input file information, and the like. Through predefined semantic grammar rules, the correctness and consistency of the information can be ensured. Thus, the related information of the simulation software can be conveniently acquired and used, and the complexity of setting and configuration is reduced.
In addition to the first two parameter sets, other parameter sets may be defined in this embodiment for setting parameters of the simulation software. Through a predefined semantic grammar rule, the input format and range of parameters can be determined, and wrong parameter combination or illegal parameter input is prevented. Therefore, the accuracy and the rationality of simulation parameters can be ensured, and the execution effect of the simulation task and the reliability of the result are improved.
In the embodiment, the simulation task is saved as the operation template, so that the defined parameter combination comprising task information, simulation software information and parameter setting information can be conveniently reused. Therefore, time and workload can be saved, and the working efficiency can be improved. While predefined semantic grammar rules may ensure the correctness of the templates so that the templates can be reliably used in multiple tasks.
The present embodiment plays a role in guiding and normalizing the conversion of simulation subtasks into computer-executable jobs according to predefined semantic grammar rules. By defining unified task information, normalized simulation software information and flexible parameter settings, the accuracy, consistency and repeatability of task execution can be improved. Meanwhile, the repeated utilization of the operation template can save time and resources.
Embodiment seven, this embodiment will be described with reference to fig. 3. The present embodiment is further defined on the method for designing a remote co-simulation platform architecture based on a super computing cloud according to the second embodiment, where the step S4 includes:
after receiving the job request sent by the server program, the super computing resource access service orders the jobs by taking the total task as a unit, and sends the jobs to the super computing job scheduling system for execution according to the job sequence;
the super-computing operation scheduling system also monitors the execution state of the task in real time in the task execution process, feeds back the execution state and log information to the server program for real-time display, and sends all simulation data path information under the task to the data synchronization server after all sub-tasks are executed.
In this embodiment, the supercomputer resource access service orders the total tasks after receiving the job request, and sends the jobs to the supercomputer job scheduling system for execution according to the job sequence. The purpose of this step is to ensure that tasks are performed in a predetermined order, avoiding conflicts or resource contention problems that may arise when concurrently performed. By performing in order, the controllability and stability of the job can be improved. The super-computing operation scheduling system monitors the execution state of the task in real time in the task execution process, and feeds back the execution state and log information to the client program for real-time display. The purpose of this step is to enable the user to know in real time the execution and progress of the task, including whether the task is running properly, whether an error has occurred, and the resources and time spent. Through real-time display, the user can take corresponding measures in time, such as parameter adjustment or job resubmission, so that the success rate and efficiency of the task are improved. And after all the sub-tasks are executed, the super-computing operation scheduling system transmits all simulation data path information under the tasks to the data synchronization server. The embodiment synchronizes and shares the data generated by simulation, ensures that the data is stored and backed up, and can be used for subsequent analysis and processing by a user. Through the synchronization and sharing of the data, the management and the utilization of the data can be conveniently carried out, and the reliability and the accessibility of the data are improved.
The method and the system ensure that the jobs are sequentially executed, and provide real-time task execution state display, and synchronization and sharing of data. These steps all help to improve the controllability, stability and operability of the task, and also provide the function of real-time tracking and management of task execution and data.
An eighth embodiment is a further limitation of the method for designing a remote co-simulation platform architecture based on a super computing cloud according to the seventh embodiment, wherein the sending the task sequence to a super computing task scheduling system for execution includes:
analyzing the operation file;
and reading a task information group, numbering the task information group according to the total task, acquiring the job documents of all the subtasks under the total task, sequencing the subtasks according to the job document numbering sequence of the subtasks, and uniformly submitting the sequencing result to a super-computing job scheduling system for execution.
In this embodiment, the job file is parsed, which includes parsing and processing the file to obtain information about the job and a task list. Through the analysis of the job file, the relevant information of each subtask, such as the number of the job document, the resources required by execution and the like, can be accurately acquired. All subtasks under the task can be determined according to the total task number in the task information group. By reading and numbering the task information groups, the association relationship between the subtasks and the total tasks can be established. This ensures that the sub-tasks are ordered and executed in a predetermined order. And sequencing the subtasks according to the serial numbers of the job documents of the subtasks. This ensures that the subtasks are submitted to the supercomputing operation scheduling system for execution in a predetermined order and avoids conflicts or resource contention problems that may be caused by concurrent execution. By uniformly submitting the subtasks, the controllability and stability of the operation can be improved.
In the embodiment, the task information group is read and numbered through analyzing the job file, and the sub-tasks are sequenced and uniformly submitted to the super-computing job scheduling system. Thus, the operation can be ensured to be executed in sequence, the conflict and the resource contention are avoided, and the controllability and the stability of the operation are improved.
The computer readable storage medium according to the ninth embodiment is used for storing a computer program, and the computer program executes the method for designing a different-place collaborative simulation platform architecture based on a super computing cloud according to any one of the second to eighth embodiments.
The computer device according to the tenth embodiment includes a memory and a processor, where the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the method for designing a different-place co-simulation platform architecture based on a super computing cloud according to any one of the second to eighth embodiments.
An eleventh embodiment is a specific example provided for the method for designing a remote co-simulation platform architecture based on a super computing cloud according to the second embodiment, and is also used for explaining the second embodiment to the eighth embodiment, specifically:
and step 1, a user with task management authority creates a collaborative simulation task by using client software, splits the simulation task into sub-tasks according to components and disciplines, and sends the sub-tasks to a simulation execution engineer.
The client software is deployed on a local work PC machine, authentication is carried out on a login user, and the user with simulation task management authority logs in the client and can establish a simulation task through a visual human-home interaction interface. When the simulation task is created, the system automatically generates a task number for the task according to the model, the specialty and the time, and the number is used as a unique identification in the simulation life cycle.
For complex multi-user collaborative simulation service scenes such as cross-component, multidisciplinary and the like, the simulation tasks are split into different subtasks according to the service scenes, and the subtasks can be issued to simulation engineers corresponding to professions and disciplines for execution.
When the subtasks are split, subtask numbers are generated for the subtasks according to the sequence of natural numbers according to the execution sequence of the subtasks and are used as the basis of the execution sequence of the subtasks. When the simulation task is issued, the task information carries the task number as the unique identifier of the task, and meanwhile, the total number of the subtasks, the number of the current subtasks and the general requirement information of the task are also contained.
And step 2, after receiving the collaborative simulation task issued by the upper level at the local client, the simulation engineer with the execution authority sets the starting and calculating parameters of the simulation software deployed on the super computing cloud at the local client, and submits the set parameters to the server program for processing.
And the execution authority simulation engineer executes the upper-level issuing simulation task through the local client interface. And selecting simulation software deployed on the super computing cloud according to task requirements, and setting a simulation software driving script and calculation parameters through a visual man-machine interaction interface.
In this step, a parameter dynamic expansion scheme is provided, and interface parameters can be increased according to the actual requirements of different simulation software, and only parameter names, parameter types, parameter values and parameter units are required to be set.
The parameter setting interfaces customized for different simulation software can be stored as parameter templates, the saved parameter templates can be imported into the client software when the simulation software is reused, and the software automatically generates the parameter setting interfaces of the simulation software according to the parameter templates.
And step 3, the server program receives the simulation task submitted in the previous step, and the simulation task is converted into a job file according to a predefined grammar rule and submitted to the super-computing resource access service.
The task request submitted by the simulation engineer client is saved as a 'group+key-value' mode job file, wherein group is a parameter group, key is a key, value is a value corresponding to the key, and one or more key-value pairs can exist under one group.
According to a predefined semantic grammar rule, the job file format and content are as follows:
the first parameter set is task information, and the key-value pair comprises a total task number, the number of subtasks under the total task and the current subtask number.
The second parameter set is simulation software information, and the key-value pair comprises a simulation software name, a version number, an installation address, driving script information and input file information.
The other parameter sets are simulation software parameter setting information, and there can be one to a plurality of parameter sets according to different software, and there can be one to a plurality of parameter key-value pairs under each parameter set.
Taking the example that the total task comprises 3 subtasks, wherein the content of a first group of parameters of the subtask 1 job file is as follows:
[Task]
TaskNO＝20230720001
TotalNumber＝3
SubNo＝1
wherein [ Task ] is a Task parameter set; the task NO is a total task number key, and 20230720001 is a total task number value; total number is the total number of subtasks key, 3 is the total number of subtasks; subtno is a subtask number key, 1 is a subtask number value.
And step 4, after receiving the job request sent in the last step, the super computing resource access service orders the jobs by taking the total task as a unit, and sends the jobs to a super computing job scheduling system for execution according to the job sequence. The execution flow is shown in fig. 3.
The super computing resource access service firstly acquires the total task number and the total number of subtasks after receiving the job submitting request sent in the last step, sorts the subtasks according to the subtask numbers after acquiring all the subtask job files, and executes the subtasks according to the sequence.
The method comprises the steps of analyzing a job file, firstly reading Task information group (corresponding to Task in the last step), then respectively reading corresponding values of a subtask total number key (corresponding to total number in the last step) and a subtask number key (corresponding to subtNo in the last step) from a Task information group according to a total Task number key (corresponding to Task in the last step), sequencing subtasks according to the sequence of the subtask numbers after the job files of all subtasks in the total Task are obtained, and uniformly submitting the subtasks to a super calculation for execution.
When computing resources on the super computing cloud are distributed, the computing resources are distributed by taking the total task as a unit, and the distributed resources belong to all the tasks before the total task is executed, so that all the subtasks under the tasks are ensured to be continuously executed according to the submitted sequence.
And in the task execution process, the task execution state is monitored in real time, and the execution state and log information are fed back to the task submitting client for real-time display. And after all the subtasks are executed, transmitting all the data path information of the subtasks to the data synchronization service.
And step 5, after receiving the task completion instruction of the super computing resource access service, the data synchronization service synchronously transfers the simulation data to the data sharing server, and performs classified storage management on the data according to the form of adding the index to the label.
And 6, the simulation task creator and the simulation execution engineer can access the simulation task data in the remote data sharing server through the local client, and can browse or download the data to the local machine at the remote server.
The task executor can access the data information of the self-submitted job in the data sharing server through the local client, and the task manager can access the data information of the self-created task in the data sharing server through the local client.
The embodiment provides a different-place collaborative simulation platform architecture design method based on a super-computing cloud, which solves the key problems of data collaboration and interaction in different-place collaborative simulation under complex simulation service scenes such as cross-component, multidisciplinary and the like in the field of aeroengines. In the simulation execution process, the artificial participation content is less, and the calculation execution condition on the far-end super-calculation cloud can be monitored in real time through the local client.
The technical solution provided by the present invention is described in further detail above with reference to the accompanying drawings, which is to highlight the advantages and benefits, not to limit the present invention, and any modification, combination of embodiments, improvement and equivalent substitution etc. within the scope of the spirit principles of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. The utility model provides a different place co-simulation platform framework based on super computing cloud which characterized in that, the frame includes:
the system comprises a client, a server, a super computing resource access server and a data sharing server;
the client comprises a task management authority client and an execution client, and is used for issuing a simulation task and transmitting the simulation task to the server;
the server is used for receiving the simulation task and uploading the simulation task to the super computing resource access server according to the task processing executable by the computer;
and the super computing resource access server receives the job to carry out resource scheduling and stores the resource scheduling data to a data sharing server.
2. The utility model provides a different place collaborative simulation platform architecture design method based on super computing cloud, which is characterized in that the method comprises the following steps:
s1: the task management authority client creates a collaborative simulation task locally, splits the collaborative simulation task into different simulation subtasks, and sends the different simulation subtasks to the execution client according to the parts and disciplines of the simulation subtasks;
s2: the execution client receives the simulation subtask, selects simulation software of the super computing cloud according to the requirement of the simulation subtask, and submits the simulation subtask to a server program deployed on a remote application server through the execution client;
s3: the method comprises the steps that a server program receives a simulation subtask submitted by an execution client, converts the simulation subtask into a job executable by a computer, and uniformly transmits the job to a super computing resource access server;
s4: the super computing resource access server receives the job, performs resource scheduling processing according to the job, and sends simulation data of the scheduling processing to the data synchronization server;
s5: and the data synchronization server receives the simulation data and synchronously transfers the simulation data to the data sharing server, and the data sharing server performs classified storage management according to the data type.
3. The method for designing the offsite co-simulation platform architecture based on the super computing cloud according to claim 2, wherein the step S1 comprises: when the subtasks are split, subtask numbers are generated for the subtasks according to the subtask execution sequence and the natural number sequence and are used as the basis of the subtask execution sequence.
4. The method for designing the remote co-simulation platform architecture based on the super computing cloud according to claim 2, wherein the client is developed by adopting a QT framework technology.
5. The method for designing the offsite co-simulation platform architecture based on the supercomputer cloud according to claim 2, wherein the step S3 of converting the simulation subtasks into computer-executable jobs comprises: and storing the simulation task into a job template in a group+key-value mode according to a predefined semantic grammar rule, wherein group is a parameter group, key is a key, and value is a value corresponding to the key.
6. The method for designing a different-place co-simulation platform architecture based on a super computing cloud according to claim 5, wherein the predefined semantic grammar rule comprises:
the first parameter group is task information, and the key-value pair comprises a total task number, the number of subtasks under the total task and the current subtask number;
the second parameter group is simulation software information, and the key-value pair comprises a simulation software name, a version number, an installation address, driving script information and input file information;
the other parameter sets are simulation software parameter setting information.
7. The method for designing the offsite co-simulation platform architecture based on the super computing cloud according to claim 2, wherein the step S4 comprises:
after receiving the job request sent by the server program, the super computing resource access service orders the jobs by taking the total task as a unit, and sends the jobs to the super computing job scheduling system for execution according to the job sequence;
the super-computing operation scheduling system also monitors the execution state of the task in real time in the task execution process, feeds back the execution state and log information to the client program for real-time display, and transmits all simulation data path information under the task to the data synchronization server after all sub-tasks are executed.
8. The method for designing a different-place co-simulation platform architecture based on a super computing cloud according to claim 7, wherein the sending the super computing operation scheduling system according to the operation sequence to execute the method comprises the following steps:
analyzing the operation file;
and reading a task information group, numbering the task information group according to the total task, acquiring the job documents of all the subtasks under the total task, sequencing the subtasks according to the job document numbering sequence of the subtasks, and uniformly submitting the sequencing result to a super-computing job scheduling system for execution.
9. A computer readable storage medium for storing a computer program for executing a method for offsite co-simulation platform architecture design based on a super computing cloud according to any one of claims 2-8.
10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes a method for designing a super computing cloud-based off-site co-simulation platform architecture according to any one of claims 2 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311298490.3A CN117390841B (en) | 2023-10-09 | 2023-10-09 | Different-place collaborative simulation platform architecture based on super computing cloud and design method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311298490.3A CN117390841B (en) | 2023-10-09 | 2023-10-09 | Different-place collaborative simulation platform architecture based on super computing cloud and design method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117390841A true CN117390841A (en) | 2024-01-12 |
| CN117390841B CN117390841B (en) | 2024-06-11 |
Family
ID=89436535
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311298490.3A Active CN117390841B (en) | 2023-10-09 | 2023-10-09 | Different-place collaborative simulation platform architecture based on super computing cloud and design method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117390841B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102306370A (en) * | 2011-08-26 | 2012-01-04 | 浙江大学 | Digital image processing system based on cloud computing |
| CN102646223A (en) * | 2012-02-17 | 2012-08-22 | 中国电力科学研究院 | Multi-user remote parallel computation system |
| CN103458022A (en) * | 2013-08-30 | 2013-12-18 | 国家电网公司 | Multi-user different-place concurrent collaborative system |
| CN103617509A (en) * | 2013-12-13 | 2014-03-05 | 国家电网公司 | Cloud computation-based power system simulation data management method |
| US20150263900A1 (en) * | 2014-03-11 | 2015-09-17 | Schlumberger Technology Corporation | High performance distributed computing environment particularly suited for reservoir modeling and simulation |
| WO2016101638A1 (en) * | 2014-12-23 | 2016-06-30 | 国家电网公司 | Operation management method for electric power system cloud simulation platform |
| CN111309491A (en) * | 2020-05-14 | 2020-06-19 | 北京并行科技股份有限公司 | Operation cooperative processing method and system |
| CN112260331A (en) * | 2020-12-21 | 2021-01-22 | 中国电力科学研究院有限公司 | Extra-high voltage alternating current-direct current power grid simulation platform and construction method |
| CN115879323A (en) * | 2023-02-02 | 2023-03-31 | 西安深信科创信息技术有限公司 | Automatic driving simulation test method, electronic device and computer readable storage medium |
| CN115906499A (en) * | 2022-12-05 | 2023-04-04 | 中国航空发动机研究院 | Heterogeneous system-oriented aircraft engine integrated simulation workflow engine system |
-
2023
- 2023-10-09 CN CN202311298490.3A patent/CN117390841B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102306370A (en) * | 2011-08-26 | 2012-01-04 | 浙江大学 | Digital image processing system based on cloud computing |
| CN102646223A (en) * | 2012-02-17 | 2012-08-22 | 中国电力科学研究院 | Multi-user remote parallel computation system |
| CN103458022A (en) * | 2013-08-30 | 2013-12-18 | 国家电网公司 | Multi-user different-place concurrent collaborative system |
| CN103617509A (en) * | 2013-12-13 | 2014-03-05 | 国家电网公司 | Cloud computation-based power system simulation data management method |
| US20150263900A1 (en) * | 2014-03-11 | 2015-09-17 | Schlumberger Technology Corporation | High performance distributed computing environment particularly suited for reservoir modeling and simulation |
| WO2016101638A1 (en) * | 2014-12-23 | 2016-06-30 | 国家电网公司 | Operation management method for electric power system cloud simulation platform |
| CN111309491A (en) * | 2020-05-14 | 2020-06-19 | 北京并行科技股份有限公司 | Operation cooperative processing method and system |
| CN112260331A (en) * | 2020-12-21 | 2021-01-22 | 中国电力科学研究院有限公司 | Extra-high voltage alternating current-direct current power grid simulation platform and construction method |
| CN115906499A (en) * | 2022-12-05 | 2023-04-04 | 中国航空发动机研究院 | Heterogeneous system-oriented aircraft engine integrated simulation workflow engine system |
| CN115879323A (en) * | 2023-02-02 | 2023-03-31 | 西安深信科创信息技术有限公司 | Automatic driving simulation test method, electronic device and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 黄博;姚烨;王佳川: "航空发动机数值仿真智能综合集成平台架构研究", 计算机集成制造系统, vol. 28, no. 7, 31 July 2022 (2022-07-31), pages 2113 - 2118 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117390841B (en) | 2024-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Scolati et al. | A Containerized Big Data Streaming Architecture for Edge Cloud Computing on Clustered Single-board Devices. | |
| US8020044B2 (en) | Distributed batch runner | |
| CN107239675A (en) | Biological information analysis system based on cloud platform | |
| CN102012840A (en) | Batch data scheduling method and system | |
| CN107016094B (en) | Project shared file multi-person collaborative development method, device and system | |
| CN104050275A (en) | Method and system for reviewing project | |
| CN110389834A (en) | A kind of method and apparatus for submitting deep learning training mission | |
| CN112256406B (en) | Operation flow platformization scheduling method | |
| CN111367804B (en) | Method for realizing front-end cooperative debugging based on cloud computing and network programming | |
| CN107944743A (en) | The operation and maintenance method, apparatus and system of a kind of large computer system | |
| CN110766375A (en) | Work platform task construction system and construction method | |
| CN112347080B (en) | Data migration method and related device | |
| CN117390841B (en) | Different-place collaborative simulation platform architecture based on super computing cloud and design method | |
| CN115794355B (en) | Task processing method, device, terminal equipment and storage medium | |
| CN111459748A (en) | Operation and maintenance control method, system and device based on command control and storage medium | |
| US11782747B2 (en) | System and method for notebook processing to handle job execution in cross-cloud environment | |
| Pan et al. | Design and Implementation of Server Management System Based on Docker | |
| CN111913791B (en) | Task scheduling method, device, equipment and computer readable storage medium | |
| CN114090242A (en) | Model training method, device, equipment, storage medium and program product | |
| CN112597121A (en) | Logic script processing method and device, electronic equipment and storage medium | |
| CN114691766A (en) | Data acquisition method and device and electronic equipment | |
| CN112115118A (en) | Database pressure measurement optimization method and device, storage medium and electronic equipment | |
| Abidi et al. | Desktop grid computing at the age of the web | |
| CN113504956B (en) | Method, device, equipment and medium for calling public function under micro service platform | |
| KR101273374B1 (en) | System and Method for executing simulation workflow based on plug-in |
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 |