Simulation resource pool management scheduling method and system and simulation method and system

By building equipment and model resource pools, calculating priority and health values, and dynamically allocating simulation equipment, the equipment hardware requirements and resource elastic expansion problems in the real-time simulation field are solved, the efficiency and stability of the simulation system are improved, and it is suitable for cloud simulation platforms. .

CN119088562APending Publication Date: 2024-12-06CHANGSHA KELIANG TECH CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411203365.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing technology is difficult to meet the equipment hardware requirements in the real-time simulation field, and cannot achieve dynamic deployment and resource elastic expansion, resulting in insufficient resource allocation efficiency and stability.

Method used

By building a device resource pool and a model resource pool, calculate the priority values ​​of the device and model, select the simulation equipment based on the optimal priority value, and judge the device availability based on the health value, and realize the dynamic allocation of simulation equipment and the management and scheduling of resource pools. .

Benefits of technology

It meets the hardware requirements of real-time simulation equipment, realizes the execution of dynamic deployment model and elastic resource expansion of users, shortens project cycles, reduces capital needs, and improves the robustness and applicable scenarios of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119088562A_ABST
    Figure CN119088562A_ABST
Patent Text Reader

Abstract

The invention discloses a simulation resource pool management scheduling method and system and a simulation method and system.The simulation resource pool management scheduling method comprises the steps that an equipment resource pool and a model resource pool are constructed, related information of multiple pieces of simulation equipment is configured in the equipment resource pool, and multiple pieces of model operation tense information are configured in the model resource pool; establishing a simulation task, and obtaining operation tense information of a model corresponding to the simulation task from the model resource pool; and calculating a priority value, matched with the model, of the simulation equipment in the equipment resource pool, calling the simulation equipment based on the optimal priority value, and allocating the called simulation equipment to the corresponding simulation task. According to the method, the equipment hardware requirement of real-time simulation can be met, and the requirements of dynamic deployment model execution and resource elastic expansion of a user are met.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of cloud simulation technology, and in particular to a simulation resource pool management and scheduling method and system, and a simulation method and system. Background Art

[0002] The rapid development of network technology has brought various cloud technologies closer to our lives. In the current real-time simulation field, there are equipment hardware requirements that cannot be dynamically deployed and elastically expanded. Resource allocation originates from various software and hardware platforms, and its implementation is mainly divided into four algorithms: FCFS, SLJ, priority scheduling, and time slice rotation. Each of the four algorithms has its specific scope of application. In practical applications, in order to better handle different types of jobs, it may be necessary to combine and adjust these algorithms to maximize the efficiency and stability of the system.

[0003] The allocation of real-time simulation resources is quite different from traditional I-layer cloud services. Therefore, it is a great challenge to allocate the real-time simulation resource pool, meet user response time, shorten model resource matching, and obtain the optimal resource selection. Summary of the invention

[0004] The purpose of the present invention is to provide a simulation resource pool management and scheduling method and system, a simulation method and system to address the deficiencies of the above-mentioned prior art, so as to meet the equipment hardware requirements of real-time simulation and meet the user's dynamic deployment model execution and resource elastic expansion requirements.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A simulation resource pool management and scheduling method is characterized by comprising the following steps:

[0007] Step 1: construct a device resource pool and a model resource pool, wherein the device resource pool is configured with relevant information of multiple simulation devices, and the model resource pool is configured with runtime information of multiple models;

[0008] Step 2: Establish a simulation task and obtain the runtime information of the model corresponding to the simulation task from the model resource pool;

[0009] Step 3, calculating the priority value of the simulation device in the device resource pool and matching the model in step 2, retrieving the simulation device based on the optimal priority value, and allocating the retrieved simulation device to the corresponding simulation task.

[0010] Furthermore, the step 3 also includes first determining whether there is the latest allocation result information for the model. If so, and the simulation device corresponding to the latest allocation result is available, directly allocating the corresponding simulation device to the corresponding simulation task; otherwise, calculating the priority value.

[0011] As a preferred method, in step 3, multiple calculation priority values ​​are based on the following information: device static configuration index, device dynamic calculation index, number of remaining available cores of the device, total number of cores of the device, total number of historical allocations of the device, total number of historical allocations of the model, total number of historical failed runs of the device, total historical model running time of the device, total number of models currently allocated to the device, and whether this device is the last device allocated to the model.

[0012] Furthermore, in step 3, after the simulation device is called based on the optimal priority value, the health value of the called simulation device is calculated, and when the calculated health value is greater than a preset health threshold, the called simulation device is assigned to the corresponding simulation task.

[0013] As a preferred method, the health value of the simulated device is calculated based on the total number of models currently allocated to the device, the number of remaining available cores of the device, the number of model running failures within the last t1 time of the device, and the number of model allocation failures within the last t2 time of the device, where t1<t2.

[0014] Furthermore, after allocating the simulation device to the simulation task, the method further includes: binding the allocated simulation device to the simulation task, and saving the latest resource allocation information of the corresponding model.

[0015] Furthermore, the simulation task is configured with a task ID, and the task ID is used as a unique identifier for globally tracking and indexing transactions related to the simulation task.

[0016] Based on the same inventive concept, the present invention also provides a simulation resource pool management and scheduling system, and the simulation resource pool management and scheduling system is used to execute the simulation resource pool management and scheduling method.

[0017] Based on the same inventive concept, the present invention also provides a simulation method, which utilizes the simulation resource pool management and scheduling method to allocate simulation devices to simulation tasks and execute the simulation tasks.

[0018] Based on the same inventive concept, the present invention also provides a simulation system, which is used to execute the simulation method.

[0019] Compared with the prior art, the present invention can meet the equipment hardware requirements of real-time simulation, and meet the user's dynamic deployment model execution and resource elastic expansion requirements; when building a cloud simulation platform, the resource pool and scheduling allocation function constructed by the present invention can dynamically meet the hardware resources required by the customer's simulation model, avoiding the construction of a simulation environment; the present invention meets the dynamic usage scenarios anytime and anywhere, making cloud simulation a readily available SAAS application, which is particularly suitable for the fields of electricity, energy, etc. When high-order simulation computing power is required, it is no longer necessary to go through complex stages such as resource procurement and system debugging, which not only reduces the project's funding needs, but also shortens the project cycle, and has great economic value. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a block diagram of an embodiment of a simulation system of the present invention.

[0021] Figure 2 The present invention is a flowchart of an embodiment of simulation resource allocation. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described below in conjunction with the drawings in the embodiment. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0023] A first aspect of an embodiment of the present invention provides a simulation resource pool management and scheduling method, which includes the following steps:

[0024] Step 1: construct a device resource pool and a model resource pool, wherein the device resource pool is configured with relevant information of multiple simulation devices, and the model resource pool is configured with runtime information of multiple models;

[0025] Step 2: Establish a simulation task and obtain the runtime information of the model corresponding to the simulation task from the model resource pool;

[0026] Step 3, calculating the priority value of the simulation device in the device resource pool and matching the model in step 2, retrieving the simulation device based on the optimal priority value, and allocating the retrieved simulation device to the corresponding simulation task.

[0027] In some preferred embodiments, step 3 also includes first determining whether there is the latest allocation result information for the model. If so, and the simulation device corresponding to the latest allocation result is available, directly allocating the corresponding simulation device to the corresponding simulation task; otherwise, calculating the priority value.

[0028] In the step 3, it is preferred but not limited to multiple calculation priority values ​​based on the following information: device static configuration index, device dynamic calculation index, number of remaining available cores of the device, total number of cores of the device, total number of historical allocations of the device, total number of historical allocations of the model, total number of historical failed runs of the device, total historical model running time of the device, total number of models currently allocated to the device, and whether the current device is the device that was last allocated to the model.

[0029] In some preferred embodiments, in step 3, after calling the simulation device based on the optimal priority value, the health value of the called simulation device is calculated, and when the calculated health value is greater than a preset health threshold, the called simulation device is assigned to the corresponding simulation task.

[0030] The health value of the simulated device is preferably calculated based on, but not limited to, the total number of models currently allocated to the device, the number of remaining available cores of the device, the number of model running failures within the last t1 time of the device, and the number of model allocation failures within the last t2 time of the device, where t1<t2.

[0031] In some preferred embodiments, after allocating the simulation device to the simulation task, the method further includes: binding the allocated simulation device to the simulation task and saving the latest resource allocation information of the corresponding model.

[0032] In some preferred embodiments, the simulation task is configured with a task ID, and the task ID serves as a unique identifier for globally tracking and indexing transactions related to the simulation task.

[0033] A second aspect of an embodiment of the present invention provides a simulation method, which utilizes the simulation resource pool management and scheduling method to allocate simulation devices to simulation tasks and execute the simulation tasks.

[0034] A third aspect of the embodiments of the present invention provides a simulation resource pool management and scheduling system, which is used to execute the simulation resource pool management and scheduling method. A fourth aspect of the embodiments of the present invention provides a simulation system, which is used to execute the simulation method.

[0035] The present invention deploys a large-scale simulation machine cluster in the cloud, uniformly performs task-based resource scheduling and allocation through a resource pool management and scheduling system, and provides users with interface-based simulation resource allocation services. It is an innovative application that leads the industry.

[0036] The present invention is based on a service interface, provides real-time simulation equipment hardware requirements, meets users' dynamic deployment model execution and resource elasticity requirements, and is an innovative application for the cloud simulation industry.

[0037] like Figure 1The present invention abstracts the simulation environment into a resource pool, which is managed uniformly by a simulation resource pool management and scheduling system 1, and achieves matching scheduling between simulation task resource requirements and resource pool devices through a matching algorithm.

[0038] The simulation resource pool management and scheduling system 1 preferably includes, but is not limited to, the following functional modules, and the functions of each module are described as follows:

[0039] 1. Device resource pool 101: Through configuration, the information of all simulation devices 201, 202, 203 in the cloud simulation machine cluster is maintained in a table. The relevant information of the simulation devices 201, 202, 203 includes the model, device IP, interface type, interface protocol, number of simulation cores, etc. of the simulation devices 201, 202, 203. The device resource pool 101 is a set of resources that can be scheduled in the current platform.

[0040] 2. Model resource pool 102: Model resource pool 102 is a dynamically constructed resource pool. When the simulation environment needs to execute a model, it needs to build a model runtime state before loading it onto the simulation devices 201, 202, and 203. Model resource pool 102 includes multiple model runtime state information, such as the runtime configuration, compilation status, executable program, and data reporting configuration information of the model file. After the simulation system is started, the model resource pool 102 is constructed. When the simulation task is initiated, the model runtime state is obtained from the model resource pool 102 and loaded onto the simulation devices 201, 202, and 203.

[0041] 3. Simulation task management module 103: used to manage the simulation process, including the initiation, resource allocation, execution process, task completion and resource recovery of simulation tasks, and also provides query of the real-time status of simulation tasks.

[0042] 4. Simulation resource monitoring and allocation module 104: Through the data collected by resource monitoring, the matching priority values ​​of the devices and models in the resource pool are dynamically calculated in real time; based on the priority values ​​and model resource pool information, resource demand matching is performed, a matching solution is generated, and a health check is performed. The algorithm flow of a preferred embodiment of simulation resource allocation is as follows: Figure 2 As shown, the following steps are included:

[0043] 4.1. Establish a simulation task and obtain the model runtime information from the model resource pool 102. The model runtime information is constructed before the simulation task starts, which generally refers to the model parameter setting, compilation, operation and debugging, etc. At the same time, a simulation task ID is generated, which will be used as the unique mark of the simulation transaction for global tracking and indexing.

[0044] 4.2. Determine whether the model has the latest allocation result information. If so, first obtain the simulation resource status corresponding to the latest allocation result. If the resource is available, use it as the optimal choice. Otherwise, enter the allocation process based on priority.

[0045] 4.3. First determine whether there are any unallocated simulation resources, that is, whether the priority list of the simulation environment matched by the model is empty. If so, the allocation fails. Otherwise, select the simulation device with the highest priority from the priority queue as the allocation target for this time.

[0046] The priority calculation formula is preferably but not limited to:

[0047]

[0048] in:

[0049] S d Represents the static configuration index of the device, which is calculated as follows

[0050] S(d)=B md +a2*C d +b2*F d

[0051] B md Represents the binding index of the model and the device configured only by the configuration module (configured by the resource configuration module 107).

[0052] F d Indicates the main frequency of the device.

[0053] D d Represents the device dynamic calculation index, which is calculated as follows

[0054] D(d)=c2*L d +d*N d

[0055] L d Represents the current system load of the device (obtained in real time by the simulation resource monitoring and allocation module 104).

[0056] N d Represents the current network bandwidth occupancy rate of the device (obtained in real time by the simulation resource monitoring and allocation module 104).

[0057] C l Indicates the number of available cores remaining on the device.

[0058] C d Indicates the total number of cores on the device.

[0059] SA d Represents the total number of historical allocations of the device.

[0060] SA m Represents the total number of historical allocations of the model.

[0061] SF d Indicates the total number of failed operations in the device history.

[0062] SR d Indicates the total historical model running time of the device.

[0063] SA dm Indicates the total number of models currently assigned to the device.

[0064] I dm Indicates whether this device is the last device assigned to this model.

[0065] a1, b1, c1, a2, b2, and c2 are all operating experience coefficients, which are determined by dynamically adjusting them during operation to obtain the optimal operating value of the system.

[0066] The above priority calculation formula fully considers the data accumulated during the business operation process, and combines the data with the formula to achieve a learning optimized allocation algorithm.

[0067] 4.4. Determine whether the health value of the simulation device is greater than the health value set by the model. The health value threshold set by the system represents the probability of stable and healthy operation after resource allocation. When the calculated health value of a simulation device is less than the preset health threshold, it is very likely to become a factor that causes the overall system to be unable to operate stably. This value is also dynamically calculated by the simulation resource monitoring and allocation module 104 through an intelligent algorithm.

[0068] The health value calculation formula is preferably but not limited to:

[0069] f(d)=SA dm +C l +a3*SF dth +b3*SF dtm

[0070] in:

[0071] SA dm Indicates the total number of models currently assigned to the device.

[0072] C l Indicates the number of available cores remaining on the device.

[0073] SF dth Indicates the number of model run failures on the device within the last t1 time (e.g., within the last hour).

[0074] SF dtm Represents the number of model allocation failures on the device within the last t2 time (such as within the last month).

[0075] a3 and b3 are both operating experience coefficients, which are determined by dynamically adjusting them during operation to obtain the optimal operating value of the system.

[0076] When calculating the health during the final check of the allocation, we not only consider whether the basic hardware requirements of the device can be met, but also the health indicators of the system to avoid allocating devices that meet the basic requirements to simulation tasks, which may cause overall system stability problems. The coefficients set based on empirical values ​​can maximize the robustness of the system.

[0077] 4.5. After successful allocation, bind the corresponding simulation device to the simulation task and save the latest resource allocation information corresponding to the model.

[0078] 4.6. The unique mark of simulation task ID will be set in the runtime environment of the model. All data reported when the model is running will carry the task ID, thereby establishing the association between the data and the simulation task, providing protection for the data isolation of each business system.

[0079] 5. Simulation service interface 105: according to the simulation service requirements sent by the user through the service systems 301, 302, 303 in the simulation service application unit 3, it provides interface services connected to the simulation resource pool management and scheduling system 1 to execute simulation tasks.

[0080] 6. Simulation device interface adaptation module 106 : used to call different drivers in the simulation device and driver unit 2 to drive the corresponding simulation devices 201 , 202 , 203 in the simulation device and driver unit 2 .

[0081] 7. Resource configuration module 107: As a tool module, it is used for static configuration, such as setting resource pools.

[0082] In summary, the present invention solves the resource allocation problem of the cloud simulation platform. By combining the adaptation of multiple simulation device interfaces, the construction of device resource pools, the construction of model resource pools, the resource matching allocation algorithm, and the resource monitoring function, it can meet the compatibility of various simulation devices and the resource allocation requirements of various simulation-based businesses, greatly improving the applicable scenarios of the simulation system and having great promotion and application value.

[0083] The priority calculation formula proposed in the present invention is a learning algorithm model, which is based on the iterative optimization results of system sample data and fully adopts various factors that affect the optimal allocation effect in actual scenarios. It can support the optimization of allocation results and has strong adaptability to the simulation resource pool.

[0084] The health calculation formula proposed in the present invention is an empirical value algorithm model that can improve the system survivability of the entire system in abnormal and dangerous situations. In actual scenarios, the overall security is often ignored for the purpose of maximizing resource utilization, resulting in large-scale jitters in the system under high load conditions until the system crashes, which is often unacceptable in cloud service scenarios. The present invention considers the overall security of the system by setting the health level and allocating within an appropriate range.

[0085] The above-mentioned device resource pool 101, model resource pool 102, simulation task management module 103, simulation resource monitoring and allocation module 104, simulation service interface 105, simulation device interface adaptation module 106, and resource configuration module 107 are used to constitute a simulation resource pool management and scheduling system 1, and then constitute a simulation system, which is used to implement the simulation resource pool management and scheduling method of the present invention and the corresponding simulation method.

[0086] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0087] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0088] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0089] In the embodiments provided by the present invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the above-described method and system embodiments are only illustrative, for example, the division of the modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0090] The units described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0091] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0092] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0093] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A simulation resource pool management and scheduling method, characterized in that: The following steps are involved: Step 1: construct a device resource pool and a model resource pool, wherein the device resource pool is configured with relevant information of multiple simulation devices, and the model resource pool is configured with runtime information of multiple models; Step 2: Establish a simulation task and obtain the runtime information of the model corresponding to the simulation task from the model resource pool; Step 3, calculating the priority value of the simulation device in the device resource pool and matching the model in step 2, retrieving the simulation device based on the optimal priority value, and allocating the retrieved simulation device to the corresponding simulation task.

2. The simulation resource pool management and scheduling method according to claim 1, characterized in that: The step 3 also includes first determining whether there is the latest allocation result information for the model. If so, and the simulation device corresponding to the latest allocation result is available, directly allocating the corresponding simulation device to the corresponding simulation task; otherwise, calculating the priority value.

3. The simulation resource pool management and scheduling method according to claim 1 or 2, characterized in that: In the step 3, multiple calculation priority values ​​are based on the following information: device static configuration index, device dynamic calculation index, number of remaining available cores of the device, total number of cores of the device, total number of historical allocations of the device, total number of historical allocations of the model, total number of historical failed runs of the device, total historical model running time of the device, total number of models currently allocated to the device, and whether the device is the last device allocated to the model.

4. The simulation resource pool management and scheduling method according to claim 1 or 2, characterized in that: In step 3, after the simulation device is retrieved based on the optimal priority value, the health value of the retrieved simulation device is calculated, and when the calculated health value is greater than a preset health threshold, the retrieved simulation device is assigned to the corresponding simulation task.

5. The simulation resource pool management and scheduling method according to claim 4, characterized in that: The health value of the simulated device is calculated based on the total number of models currently allocated to the device, the number of remaining available cores of the device, the number of model running failures within the device's recent t1 period, and the number of model allocation failures within the device's recent t2 period, where t1<t2.

6. The simulation resource pool management and scheduling method according to claim 1 or 2, characterized in that: After the simulation device is assigned to the simulation task, the method further includes: binding the assigned simulation device to the simulation task and saving the latest resource allocation information of the corresponding model.

7. The simulation resource pool management and scheduling method according to claim 1 or 2, characterized in that: The simulation task is configured with a task ID, and the task ID is used as a unique identifier for global tracking and indexing of simulation task related transactions.

8. A simulation resource pool management and scheduling system, characterized in that: The simulation resource pool management and scheduling system is used to execute the simulation resource pool management and scheduling method as described in any one of claims 1 to 7.

9. A simulation method, characterized in that: The simulation method uses the simulation resource pool management and scheduling method as described in any one of claims 1 to 7 to allocate simulation devices to simulation tasks and execute the simulation tasks.

10. A simulation system, characterized in that: The simulation system is used to execute the simulation method as claimed in claim 9.

Citation Information

Cited By

  • Semi-physical simulation resource scheduling method and device and storage medium

    CN120104288A

  • A scheduling method, device and storage medium for semi-physical simulation resources

    CN120104288B