Simulation resource scheduling method and device, computer equipment and storage medium

By dynamically adjusting the priority and allocation of simulation tasks in the node cluster, the problem of low efficiency in chip design simulation is solved, and efficient parallel processing of simulation tasks and fast execution of high-priority tasks are achieved.

CN118939428APending Publication Date: 2024-11-12SHANDONG YUNHAI GUOCHUANG CLOUD COMPUTING EQUIP IND INNOVATION CENT CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411073735.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-06
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

In current chip design simulation processes, there is a high degree of reliance on manual operation, making it impossible to run simulation tasks for multiple layout files simultaneously, resulting in low simulation efficiency.

Method used

By dynamically adjusting the priority and allocation of simulation tasks in the node cluster, high-priority tasks can be executed by nodes with idle or low-priority tasks, thus achieving parallel processing and efficient scheduling of simulation tasks.

Benefits of technology

It improves the efficiency of chip design simulation, reduces reliance on manual intervention, ensures the rapid completion of high-priority tasks, and enhances the overall execution efficiency of simulation tasks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118939428A_ABST
    Figure CN118939428A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of computers, and discloses a simulation resource scheduling method and device, computer equipment and a storage medium, and the simulation resource scheduling method comprises the steps: obtaining a to-be-simulated task corresponding to a target chip, and distributing the to-be-simulated task to a corresponding task queue according to the priority of the to-be-simulated task; determining the to-be-simulated task meeting the priority condition in the task queue as a target simulation task; a target node meeting the execution condition is determined in the node cluster, the target simulation task is distributed to the target node, the target simulation task is executed based on the target node, and the running state of the target node is an idle state or an occupied state. According to the method, the execution efficiency of the simulation task can be improved through deployment and scheduling of the node cluster, the manual dependency degree is reduced, and then the chip design simulation efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a simulation resource scheduling method, device, computer equipment and storage medium. Background Art

[0002] In the process of chip design, the simulation of integrated circuits for chips is usually crucial. For example, in relevant simulation schemes, the simulation of integrated circuits is mainly carried out with the help of microwave simulation software such as HFSS (High Frequency Structure Simulator) and Sonnet. HFSS is based on the finite element method and can simulate the high-frequency electromagnetic field of any three-dimensional passive structure, such as antennas, waveguides, filters, couplers, connectors and resonant cavities, and directly obtain the port characteristic impedance, transmission constant, eigenmode or resonance solution, electromagnetic field distribution diagram and other results. However, both HFSS and Sonnet simulation software are based on single-node operation, and require more manual participation.

[0003] Specifically, the grid division of the integrated circuit layout can be adjusted manually, and the split layout can be handed over to HFSS or Sonnet simulation software for finite element decomposition, frequency domain analysis, etc., and then manually analyzed according to the simulation results, and then the layout design plan is adjusted, and then the grid division, frequency domain analysis, etc. are performed again, and this cycle is repeated until the simulation results meet the design parameters. However, due to the high degree of manual dependence, this simulation method cannot run simulation tasks of multiple layout files at the same time, which leads to low chip design simulation efficiency. Summary of the invention

[0004] In view of this, the present invention provides a simulation resource scheduling method, apparatus, computer equipment and storage medium to solve the problem of low chip design simulation efficiency.

[0005] In a first aspect, the present invention provides a simulation resource scheduling method, the method comprising:

[0006] Obtain the tasks to be simulated corresponding to the target chip, and assign the tasks to be simulated to the corresponding task queues according to their priorities;

[0007] Determine the task to be simulated that meets the priority condition in the task queue as the target simulation task;

[0008] A target node that meets the execution condition is determined in the node cluster, and a target simulation task is allocated to the target node, so as to execute the target simulation task based on the target node, wherein the running state of the target node is an idle state or an occupied state.

[0009] In an optional implementation, determining a target node that meets a priority condition in a node cluster and assigning a target simulation task to the target node includes:

[0010] Determine whether the node cluster contains an idle node whose occupation state is idle;

[0011] If so, the idle node is determined as the target node, and the target simulation task is assigned to the target node;

[0012] If not, determine in the node cluster that the simulation task being executed has a lower priority than the occupied nodes of the target simulation task;

[0013] Acquire preset association information, wherein the preset association information is used to indicate an association relationship between a priority of a simulation task and a corresponding total execution cycle;

[0014] Based on the preset association information, a target total execution cycle corresponding to the priority of the simulation task in the occupied node is obtained;

[0015] Determine the target node whose simulation task execution time exceeds the target total execution period among the occupied nodes;

[0016] Terminate the simulation task being executed in the target node and assign the target simulation task to the target node.

[0017] In an embodiment of the present invention, the parallelism of simulation task execution can be dynamically adjusted according to the degree of tension of simulation resources in the node cluster to improve the execution efficiency of simulation tasks. When simulation resources are tight, high-priority simulation tasks are executed first so that tasks with short execution cycles can be completed quickly, thereby improving the simulation efficiency of the entire chip.

[0018] In an optional embodiment, the method further includes:

[0019] After terminating the simulation task being executed in the target node, the simulation task is determined as a task to be simulated, and the priority of the task to be simulated is reduced to perform the step of allocating the task to be simulated to a corresponding task queue according to the priority of the task to be simulated.

[0020] In an embodiment of the present invention, the parallelism of simulation task execution can be dynamically adjusted according to the degree of tension of simulation resources in the node cluster to improve the execution efficiency of simulation tasks. When simulation resources are tight, high-priority simulation tasks are executed first so that tasks with short execution cycles can be completed quickly, thereby improving the simulation efficiency of the entire chip.

[0021] In an optional implementation, determining a task to be simulated that satisfies a preset priority in a task queue as a target simulation task includes:

[0022] When the node cluster does not include an idle node whose occupation state is an idle state, obtaining the first priority of the simulation task executed in the node cluster;

[0023] The tasks to be simulated with a priority higher than the first priority in the task queue are determined as target simulation tasks.

[0024] In an embodiment of the present invention, when simulation resources in a node cluster are in short supply, a task to be simulated in a task queue with a priority higher than the first priority of the simulation tasks executed in the node cluster can be determined as a target simulation task to ensure that tasks with short execution cycles can be quickly executed when simulation resources are in short supply, thereby improving the simulation efficiency of the chip as a whole.

[0025] In an optional implementation, obtaining the tasks to be simulated corresponding to the target chip, and assigning the tasks to be simulated to the corresponding task queues according to the priorities of the tasks to be simulated, includes:

[0026] Estimate the execution cycle of the task to be simulated;

[0027] The priority of the task to be simulated is set according to the execution cycle, and the task to be simulated is assigned to the corresponding task queue according to the priority.

[0028] In an embodiment of the present invention, the priority of the task to be simulated can be set according to the execution cycle of the task to be simulated. Specifically, the higher the priority, the shorter the execution cycle of the task to be simulated. Conversely, the lower the priority, the longer the execution cycle of the task to be simulated, thereby ensuring that tasks with short execution cycles can be executed quickly, thereby improving the simulation efficiency of the chip as a whole.

[0029] In an optional implementation, setting the priority of the task to be simulated according to the execution cycle includes:

[0030] Obtaining an execution time interval pre-set for each priority level, wherein the length of the execution time interval is inversely proportional to the height of the priority level;

[0031] The priority of the task to be simulated is set based on the execution time interval to which the execution cycle belongs.

[0032] In an embodiment of the present invention, the priority of the task to be simulated can be set according to the execution cycle of the task to be simulated. Specifically, the higher the priority, the shorter the execution cycle of the task to be simulated. Conversely, the lower the priority, the longer the execution cycle of the task to be simulated, thereby ensuring that tasks with short execution cycles can be executed quickly, thereby improving the simulation efficiency of the chip as a whole.

[0033] In an optional embodiment, the method further includes:

[0034] Based on the preset functional modules, the layout corresponding to the target chip is divided to obtain a sub-layout;

[0035] Based on the design parameters of the sub-layout, the tasks to be simulated corresponding to the sub-layout are generated.

[0036] In an embodiment of the present invention, the layout corresponding to the target chip can be divided to obtain sub-layouts, and corresponding tasks to be simulated can be generated for the sub-layouts respectively, so as to perform distributed scheduling of the tasks to be simulated. If the simulation resources of the node cluster are sufficient, the simulation tasks of the target chip will run in parallel in each node without affecting each other, thereby improving the execution efficiency of the simulation tasks.

[0037] In an optional embodiment, the method further includes:

[0038] When the execution result of the task to be simulated does not meet the design parameters, the component parameters in the sub-layout of the task to be simulated are adjusted, and the task to be simulated is allocated to a preset position of the corresponding task queue according to the adjusted priority of the task to be simulated.

[0039] In an embodiment of the present invention, the layout corresponding to the target chip can be divided to obtain sub-layouts, and corresponding tasks to be simulated can be generated for the sub-layouts respectively, so as to perform distributed scheduling of the tasks to be simulated. If the simulation resources of the node cluster are sufficient, the simulation tasks of the target chip will run in parallel in each node without affecting each other, thereby improving the execution efficiency of the simulation tasks.

[0040] In an optional embodiment, the method further includes:

[0041] Monitor the waiting time of tasks to be simulated in the task queue;

[0042] When the waiting time exceeds a preset upper limit value corresponding to the priority of the task to be simulated, the priority of the task to be simulated is increased to a second priority;

[0043] Allocate the task to be simulated to the task queue corresponding to the second priority level.

[0044] In an embodiment of the present invention, taking into account the case where tasks to be simulated corresponding to multiple chips are executed simultaneously, due to the mechanism of giving priority to the execution of high-priority tasks, a situation may occur where low-priority tasks to be simulated are not scheduled. Therefore, for tasks to be simulated in the task queue, the priority of the task to be simulated can be dynamically increased as the waiting time for the waiting call increases, so as to avoid the situation where low-priority tasks to be simulated are not scheduled.

[0045] In a second aspect, the present invention provides a simulation resource scheduling device, the device comprising:

[0046] An acquisition module is used to acquire the tasks to be simulated corresponding to the target chip, and assign the tasks to be simulated to the corresponding task queues according to the priorities of the tasks to be simulated;

[0047] A determination module, used for determining the to-be-simulated tasks that meet the priority conditions in the task queue as target simulation tasks;

[0048] The allocation module is used to determine the target node that meets the priority condition in the node cluster and allocate the target simulation task to the target node to execute the target simulation task based on the target node, wherein the operating state of the target node is an idle state or an occupied state.

[0049] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the simulation resource scheduling method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0050] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the simulation resource scheduling method of the first aspect or any corresponding embodiment thereof.

[0051] In a fifth aspect, the present invention provides a computer program product, including computer instructions, which are used to enable a computer to execute the simulation resource scheduling method of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 is a flow chart of a simulation resource scheduling method according to an embodiment of the present invention;

[0054] Figure 2 is a flow chart of another simulation resource scheduling method according to an embodiment of the present invention;

[0055] Figure 3 It is a schematic diagram of the execution time intervals corresponding to each priority level;

[0056] Figure 4 is a flow chart of another simulation resource scheduling method according to an embodiment of the present invention;

[0057] Figure 5 It is a schematic diagram of the task scheduling process of the first-in-first-out task queue;

[0058] Figure 6 is a timing diagram of the simulation resource scheduling method provided in this embodiment;

[0059] Figure 7 is a structural block diagram of a simulation resource scheduling device according to an embodiment of the present invention;

[0060] Figure 8 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0062] In conjunction with the application scenarios on which the execution of the simulation resource scheduling method depends, the application scenarios are described here.

[0063] In the process of chip design, simulation of integrated circuits is usually crucial, and chip design simulation mainly relies on microwave simulation software such as HFSS (High Frequency Structure Simulator) and Sonnet. HFSS is based on the finite element method and can simulate the high-frequency electromagnetic field of any three-dimensional passive structure, such as antennas, waveguides, filters, couplers, connectors and resonant cavities, and directly obtain the port characteristic impedance, transmission constant, eigenmode or resonance solution, electromagnetic field distribution diagram and other results. However, both HFSS and Sonnet simulation software are based on single-node operation, and require more manual work.

[0064] Specifically, you can manually adjust the mesh division, hand over the split layout to HFSS or Sonnet simulation software for finite element decomposition, frequency domain analysis, etc., then manually analyze according to the simulation results, and then adjust the layout design plan, and then perform mesh division, frequency domain analysis, etc., and repeat this cycle until the simulation results meet the design parameters. However, due to the high degree of manual dependence, this simulation method cannot run simulation tasks of multiple layout files at the same time, which leads to low chip design simulation efficiency.

[0065] Based on this, an embodiment of the present invention provides a simulation resource scheduling method, which can first obtain the tasks to be simulated corresponding to the target chip, and assign the tasks to be simulated to the corresponding task queue according to the priority of the tasks to be simulated. Then, the tasks to be simulated that meet the priority conditions in the task queue are determined as target simulation tasks, and the target nodes that meet the priority conditions are determined in the node cluster, and then the target simulation tasks are assigned to the target nodes to execute the target simulation tasks based on the target nodes, wherein the running state of the target nodes is an idle state or an occupied state, thereby improving the execution efficiency of the simulation tasks and reducing the manual dependence through the deployment and scheduling of the node cluster, thereby improving the chip design simulation efficiency.

[0066] According to an embodiment of the present invention, an embodiment of a simulation resource scheduling method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0067] In this embodiment, a simulation resource scheduling method is provided, which can be used on the server side. Figure 1 is a flow chart of a simulation resource scheduling method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0068] Step S101 , obtaining tasks to be simulated corresponding to the target chip, and assigning the tasks to be simulated to corresponding task queues according to their priorities.

[0069] In the embodiment of the present invention, the task to be simulated may be a task for simulating an integrated circuit of a target chip. Here, the task to be simulated may be simulated by a preset simulation software, for example, the preset simulation software may include the above-mentioned HFSS or Sonnet simulation software.

[0070] After obtaining the task to be simulated corresponding to the target chip, the task to be simulated can be allocated to the corresponding priority task queue according to the priority set for the simulation task. Here, the task queue can be generated according to the preset priority.

[0071] For example, if the preset priorities include: A, B, C, D in descending order, then the corresponding generated task queues may include: task queue A, task queue B, task queue C, task queue D.

[0072] Step S102: determine the to-be-simulated task in the task queue that meets the priority condition as the target simulation task.

[0073] In an embodiment of the present invention, when there are multiple tasks to be simulated in the task queue, the task with the highest priority among the tasks to be simulated can be determined as the target simulation task that meets the priority condition. Here, if resources are scarce in the node cluster, for example, there are no idle nodes, then if the priorities of the simulation tasks being executed are not higher than the target simulation task, the following step S203 can be performed, and if the priorities of the simulation tasks being executed in the node cluster are higher than the target simulation task, then the target simulation task can wait for calling in the node cluster until the resources of the node cluster are idle.

[0074] Step S103, determining a target node that meets the execution condition in the node cluster, and assigning the target simulation task to the target node, so as to execute the target simulation task based on the target node, wherein the running state of the target node is an idle state or an occupied state.

[0075] In an embodiment of the present invention, the node cluster can be a distributed resource cluster, which includes multiple server nodes, and each server can include the above-mentioned preset simulation software. Specifically, when HFSS or Sonnet software is included in the preset simulation software, HFSS or Sonnet software can be encapsulated as a service interface that can be called. Since HFSS and Sonnet software are both software based on the windows system, they can be encapsulated in the form of windows services, and the encapsulated preset simulation software is integrated into the layout design tool. Then, the layout design tool can be integrated into the server node to automatically perform the simulation task through the server node.

[0076] In addition, any server node in the node cluster may be pre-selected as a scheduling node, and the calling node may be used to schedule the task to be simulated to a task queue and allocate the target task to the target node.

[0077] When determining the target node, if resources in the node cluster are scarce, a server node with a lower priority than the target task in the simulation task executed in the node cluster can be determined as meeting the execution condition. If resources in the node cluster are not scarce, a server node in the node cluster that is idle can be determined as meeting the execution condition.

[0078] It can be known from the above description that in an embodiment of the present invention, first, the task to be simulated corresponding to the target chip can be obtained, and the task to be simulated can be assigned to the corresponding task queue according to the priority of the task to be simulated. Then, the task to be simulated that meets the priority condition in the task queue is determined as the target simulation task, and the target node that meets the priority condition is determined in the node cluster, and then the target simulation task is assigned to the target node to execute the target simulation task based on the target node, wherein the running state of the target node is an idle state or an occupied state, thereby improving the execution efficiency of the simulation task and reducing the manual dependence through the deployment and scheduling of the node cluster, thereby improving the chip design simulation efficiency.

[0079] In this embodiment, another simulation resource scheduling method is provided, which can be used for the above-mentioned server. Figure 2 is a flow chart of another simulation resource scheduling method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0080] Step S201 : acquiring tasks to be simulated corresponding to the target chip, and assigning the tasks to be simulated to corresponding task queues according to their priorities.

[0081] Specifically, the above step S201 includes:

[0082] Step S2011, estimating the execution cycle of the task to be simulated.

[0083] Step S2012, setting the priority of the task to be simulated according to the execution cycle, and assigning the task to be simulated to the corresponding task queue according to the priority.

[0084] In the embodiment of the present invention, the execution cycle is used to indicate the simulation time consumption of the task to be simulated. Here, considering that the amount of simulation tasks corresponding to the chip is often large, when executing the simulation task, it should be ensured that the task with a short execution cycle can be executed quickly, thereby improving the simulation efficiency of the chip as a whole. Therefore, the execution cycle of the task to be simulated can be estimated, and the priority of the task to be simulated can be set according to the execution cycle.

[0085] Specifically, the priority of the task to be simulated is set according to the execution cycle, including:

[0086] Step a1, obtaining the execution time interval pre-set for each priority level, wherein the length of the execution time interval is inversely proportional to the height of the priority level.

[0087] Step a2, setting the priority of the task to be simulated based on the execution time interval to which the execution cycle belongs.

[0088] In the embodiment of the present invention, the corresponding execution time interval can be set in advance for each priority. Here, the above-mentioned priorities A, B, C, and D are taken as examples. It can be seen from the above that the priority order is A>B>C>D. Therefore, the execution time intervals of priority AD increase step by step. For example, the execution time interval 1 corresponding to priority A is 0-1 days, the execution time interval 2 corresponding to priority B is 1-2 days, the execution time interval 3 corresponding to priority C is 2-3 days, and the execution time interval 4 corresponding to priority D is 3-4 days. Figure 3 It is a schematic diagram of the execution time intervals corresponding to various priorities, wherein the tasks to be simulated in the task queue corresponding to the high priority can be called preferentially.

[0089] After estimating the execution cycle of the task to be simulated, the execution time interval to which the cycle belongs can be determined, and the priority corresponding to the execution interval is determined as the priority of the task to be simulated. For example, if the execution cycle of the task to be executed is 10 hours, the execution time interval corresponding to the task to be executed is the above-mentioned execution time interval 1, so the priority of the task to be executed can be set to level A.

[0090] Step S202: Determine the task to be simulated that meets the priority condition in the task queue as the target simulation task. Figure 1 Step S102 of the illustrated embodiment will not be described in detail here.

[0091] Step S203, determine the target node that meets the execution condition in the node cluster, and assign the target simulation task to the target node, so as to execute the target simulation task based on the target node, wherein the running state of the target node is idle state or occupied state. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0092] In an embodiment of the present invention, the priority of the task to be simulated can be set according to the execution cycle of the task to be simulated. Specifically, the higher the priority, the shorter the execution cycle of the task to be simulated. Conversely, the lower the priority, the longer the execution cycle of the task to be simulated, thereby ensuring that tasks with short execution cycles can be executed quickly, thereby improving the simulation efficiency of the chip as a whole.

[0093] In some optional embodiments, the above Figure 1 The corresponding embodiments also include:

[0094] Step S11, based on the preset functional modules, the layout corresponding to the target chip is divided to obtain sub-layouts.

[0095] Step S12, generating a task to be simulated corresponding to the sub-layout based on the design parameters of the sub-layout.

[0096] In the embodiment of the present invention, since the integrated circuit in the chip is often relatively complex, the layout of the integrated circuit can be split to obtain multiple sub-layouts, and each sub-layout can be simulated separately.

[0097] In a specific implementation, the layout may be divided based on preset functional modules, wherein the preset functional modules may be sub-circuits including preset component types, for example, the preset component types may be capacitors and resonant cavities.

[0098] When generating tasks to be simulated for the above sub-layout, design parameters of the sub-layout can be determined based on the functional expectations of the preset functional module and the preset component types, and the design parameters can be used as simulation expectations to generate tasks to be simulated.

[0099] It should be understood that when the execution result of the task to be simulated does not meet the design parameters, the component parameters in the sub-layout of the task to be simulated are adjusted, and the task to be simulated is assigned to the preset position of the corresponding task queue according to the adjusted priority of the task to be simulated.

[0100] In an embodiment of the present invention, if the execution result of the task to be simulated does not meet the design parameters, it is considered that the performance of the sub-layout corresponding to the task to be simulated does not meet the design expectations. Therefore, the component parameters in the sub-layout can be adjusted, for example, the capacity parameters of the capacitor can be adjusted.

[0101] Next, the task to be simulated can be adjusted according to the adjusted sub-layout and the original design parameters, and the original priority of the task to be simulated is maintained unchanged, so that the task to be simulated is allocated to the preset position of the original task queue.

[0102] Specifically, in the present invention, the task queue can be a first-in-first-out queue, so the preset position can be the end of the original task queue, so as to queue up and wait for the scheduling of the scheduling node according to the first-come-first-served scheduling algorithm, thereby avoiding interrupting the task execution order. Figure 5 The figure shows a schematic diagram of the task scheduling process of the first-in-first-out task queue, wherein, when executing task allocation, the pending task can be added to the end of the task queue, and the scheduling node can schedule the target processing task from the head of the queue to the running node of the node cluster for simulation. In addition, the non-exclusive mode can be used in the task queue to avoid the problem of task execution interruption.

[0103] In an embodiment of the present invention, the layout corresponding to the target chip can be divided to obtain sub-layouts, and corresponding tasks to be simulated can be generated for the sub-layouts respectively, so as to perform distributed scheduling of the tasks to be simulated. If the simulation resources of the node cluster are sufficient, the simulation tasks of the target chip will run in parallel in each node without affecting each other, thereby improving the execution efficiency of the simulation tasks.

[0104] In this embodiment, another simulation resource scheduling method is provided, which can be used for the above-mentioned server. Figure 4 is a flowchart of another simulation resource scheduling method according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0105] Step S401 : acquiring tasks to be simulated corresponding to the target chip, and assigning the tasks to be simulated to corresponding task queues according to their priorities.

[0106] Step S402: determine the to-be-simulated task in the task queue that meets the priority condition as the target simulation task.

[0107] Specifically, the above step S402 includes:

[0108] Step S4021, when the node cluster does not include an idle node whose occupation state is an idle state, obtain the first priority of the simulation task executed in the node cluster.

[0109] Step S4022: Determine the to-be-simulated task with a priority higher than the first priority in the task queue as the target simulation task.

[0110] In the embodiment of the present invention, when the node cluster does not include an idle node whose occupation state is an idle state, it can be considered that the simulation resources in the node cluster are in short supply. At this time, high-priority short-period simulation tasks in the task queue can be executed preferentially.

[0111] Specifically, when simulation resources in the node cluster are in short supply, the first priority can be determined first. Here, the priorities of the simulation tasks being executed in the node cluster can be counted, and the lowest priority can be determined as the first priority. Then, the task allocation in the task queue with a priority higher than the first priority is monitored in real time. When it is detected that a task to be simulated is newly added to the task queue, the task to be simulated can be determined as the target simulation task.

[0112] Step S403, determine the target node that meets the execution condition in the node cluster, and assign the target simulation task to the target node, so as to execute the target simulation task based on the target node, wherein the running state of the target node is idle state or occupied state. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0113] In an embodiment of the present invention, when simulation resources in a node cluster are in short supply, a task to be simulated in a task queue with a priority higher than the first priority of the simulation tasks executed in the node cluster can be determined as a target simulation task to ensure that tasks with short execution cycles can be quickly executed when simulation resources are in short supply, thereby improving the simulation efficiency of the chip as a whole.

[0114] In some optional implementations, the above step S101 includes:

[0115] Step S21, determining whether the node cluster includes an idle node whose occupation state is an idle state; if so, executing step S22, if not, executing step S23.

[0116] Step S22: determine the idle node as the target node, and assign the target simulation task to the target node.

[0117] Step S24, determining in the node cluster that the priority of the executed simulation task is lower than the occupied nodes of the target simulation task.

[0118] Step S25, obtaining preset association information, wherein the preset association information is used to indicate the association relationship between the priority of the simulation task and the corresponding total execution cycle.

[0119] Step S26, based on the preset association information, obtaining the target total execution cycle corresponding to the priority of the simulation task in the occupied node.

[0120] Step S23, determining the target node whose simulation task execution time exceeds the target total execution period among the occupied nodes.

[0121] Step S23, terminate the simulation task being executed in the target node, and allocate the target free task to the target node.

[0122] In the embodiment of the present invention, it is firstly determined whether the simulation resources of the node cluster are tight, that is, whether the node cluster contains an idle node whose occupation state is idle. If so, the idle node can be directly determined as the target node to ensure the parallel execution of the simulation task as much as possible and improve the execution efficiency. If not, it is ensured that the high-priority short-period simulation task in the task queue is executed first.

[0123] In specific implementation, first, the occupied node of the executed simulation task with a lower priority than the target simulation task can be determined in the node cluster. If no such node exists, the target simulation task is indicated to wait for invocation. If so, the target simulation task is scheduled to the target node for execution through the scheduling node.

[0124] When determining the target node, the node with execution timeout can be determined in the occupied nodes. Specifically, the preset association information can be first obtained to query the target total execution cycle corresponding to the priority of the simulation task executed in the occupied node according to the association information. Here, the target total execution cycle is the upper limit value of the execution time corresponding to the priority of the simulation task executed in the occupied node. When the execution time of the simulation task exceeds the target total execution cycle, the occupied node can be determined as a node with execution timeout.

[0125] It should be understood that the corresponding total execution cycle can be set in advance for each level of priority, wherein the total execution cycle is used to indicate the upper limit of the execution time of the simulation task under the priority. Here, the total execution cycle can be determined according to the execution time interval. Specifically, taking the above priorities A, B, C, and D as examples, it can be seen from the above that the priority order is A>B>C>D, and the execution time interval 1 corresponding to priority A is 0-1 days, the execution time interval 2 corresponding to priority B is 1-2 days, the execution time interval 3 corresponding to priority C is 2-3 days, and the execution time interval 4 corresponding to priority D is 3-4 days.

[0126] Then, the total execution period corresponding to each priority level can exceed the execution time interval corresponding to the priority level. The total execution period 1 corresponding to priority A is more than 1 day, the total execution period 2 corresponding to priority B is more than 2 days, the total execution period 3 corresponding to priority C is more than 3 days, and the total execution period 4 corresponding to priority D is more than 4 days.

[0127] In addition, if the occupied nodes do not include a timeout node, the node with the lowest priority of the simulation task executed in the occupied nodes may be determined as the target node to ensure that the simulation task with a high priority can be executed first.

[0128] In order to improve the credibility of the simulation results, after the target node is determined, the simulation task being executed in the target node can be terminated and the priority of the task to be simulated can be lowered to execute the step of allocating the task to be simulated to the corresponding task queue according to the priority of the task to be simulated in step S101.

[0129] Specifically, when executing the above step S101 for the task to be simulated, the priority of the task to be simulated can be reduced by one level, and the task to be simulated can be assigned to the end of the corresponding priority task queue to wait for call, thereby achieving orderly execution of the task to be simulated while ensuring the confidence of the simulation result of the task to be simulated.

[0130] It should be understood that if the priority of the task to be simulated is the lowest and cannot be downgraded, the original priority of the task to be simulated can be maintained to assign the task to be simulated to the end of the original task queue to wait for invocation.

[0131] In an embodiment of the present invention, the parallelism of simulation task execution can be dynamically adjusted according to the degree of tension of simulation resources in the node cluster to improve the execution efficiency of simulation tasks. When simulation resources are tight, high-priority simulation tasks are executed first so that tasks with short execution cycles can be completed quickly, thereby improving the simulation efficiency of the entire chip.

[0132] In some optional embodiments, the above Figure 1 The corresponding embodiments also include:

[0133] Step S31, monitoring the waiting time of the tasks to be simulated in the task queue.

[0134] Step S32: when the waiting time exceeds a preset upper limit value corresponding to the priority of the task to be simulated, the priority of the task to be simulated is increased to a second priority.

[0135] Step S33: assigning the task to be simulated to the task queue corresponding to the second priority.

[0136] In the embodiment of the present invention, it is considered that when tasks to be simulated corresponding to multiple chips are executed simultaneously, due to the mechanism of giving priority to the execution of tasks with high priority, there may be a situation where tasks to be simulated with low priority are never scheduled.

[0137] Therefore, for the task to be simulated in the task queue, the priority of the task to be simulated can be dynamically increased as the waiting time of the waiting call increases. Specifically, a corresponding preset upper limit value can be set for each level of priority, and when the waiting time of the task to be simulated in the task queue corresponding to the priority exceeds the corresponding preset upper limit value, the priority of the task to be simulated is increased to a second priority, wherein the second priority is at least one level higher than the current priority.

[0138] Next, the above-mentioned task to be simulated can be assigned to the end of the task queue corresponding to the second priority and wait for calling. It should be understood that if the waiting time of the task to be simulated exceeds the corresponding preset upper limit value of the second priority queue again, the priority of the task to be simulated can be increased again until the task to be simulated is executed.

[0139] In addition, in addition to increasing the priority of the low priority simulation task, the priority of the low priority simulation task can also be reduced. Specifically, the number of high priority tasks can be counted, and if the number exceeds the upper limit, the priority of the low priority simulation task can be reduced to ensure the execution efficiency of the high priority simulation task.

[0140] In an embodiment of the present invention, taking into account the case where tasks to be simulated corresponding to multiple chips are executed simultaneously, due to the mechanism of giving priority to the execution of high-priority tasks, a situation may occur where low-priority tasks to be simulated are not scheduled. Therefore, for tasks to be simulated in the task queue, the priority of the task to be simulated can be dynamically increased as the waiting time for the waiting call increases, so as to avoid the situation where low-priority tasks to be simulated are not scheduled.

[0141] like Figure 6Shown is a timing diagram of the simulation resource scheduling method provided in this embodiment, wherein the layout design tool can be used to divide the layout corresponding to the target chip based on preset functional modules to obtain sub-layouts, generate tasks to be simulated corresponding to the sub-layouts based on design parameters of the sub-layouts, and submit the tasks to be simulated to the scheduling node.

[0142] Then, the scheduling node can assign the tasks to be simulated to the corresponding task queues according to their priorities, and schedule the target simulation tasks in the task queues to the target nodes in the node cluster for execution, store the execution results in the database, and delete the completed tasks from the task queues.

[0143] Next, the database can respond to the query request of the layout design tool for the execution result and return the query result. It should be understood that if the execution result does not match the design parameter, the scheduling node can adjust the component parameters in the sub-layout of the task to be simulated, and assign the task to be simulated to the preset position of the corresponding task queue according to the adjusted priority of the task to be simulated; if it matches, it is determined that the simulation of the task to be simulated is successful.

[0144] In summary, in an embodiment of the present invention, first, the task to be simulated corresponding to the target chip can be obtained, and the task to be simulated can be assigned to the corresponding task queue according to the priority of the task to be simulated. Then, the task to be simulated that meets the priority condition in the task queue is determined as the target simulation task, and the target node that meets the priority condition is determined in the node cluster, and then the target simulation task is assigned to the target node to execute the target simulation task based on the target node, wherein the running state of the target node is an idle state or an occupied state, thereby improving the execution efficiency of the simulation task and reducing the manual dependence through the deployment and scheduling of the node cluster, thereby improving the chip design simulation efficiency.

[0145] In the present embodiment, a simulation resource scheduling device is also provided, and the device is used to implement the above-mentioned embodiment and preferred implementation mode, and the description has been made no further. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented with software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.

[0146] This embodiment provides a simulation resource scheduling device, such as Figure 7 As shown, including:

[0147] The acquisition module 701 is used to acquire the tasks to be simulated corresponding to the target chip, and allocate the tasks to be simulated to the corresponding task queues according to the priorities of the tasks to be simulated;

[0148] A determination module 702 is used to determine the to-be-simulated task that satisfies the priority condition in the task queue as a target simulation task;

[0149] The allocation module 703 is used to determine the target node that meets the priority condition in the node cluster, and allocate the target simulation task to the target node to execute the target simulation task based on the target node, wherein the operating state of the target node is an idle state or an occupied state.

[0150] In some optional implementations, the allocation module 703 is further configured to:

[0151] Determine whether the node cluster contains an idle node whose occupation state is idle;

[0152] If so, the idle node is determined as the target node, and the target simulation task is assigned to the target node;

[0153] If not, determine in the node cluster that the simulation task being executed has a lower priority than the occupied nodes of the target simulation task;

[0154] Acquire preset association information, wherein the preset association information is used to indicate an association relationship between a priority of a simulation task and a corresponding total execution cycle;

[0155] Based on the preset association information, a target total execution cycle corresponding to the priority of the simulation task in the occupied node is obtained;

[0156] Determine the target node whose simulation task execution time exceeds the target total execution period among the occupied nodes;

[0157] Terminate the simulation task being executed in the target node and assign the target simulation task to the target node.

[0158] In some optional implementations, the allocation module 703 is further configured to:

[0159] After terminating the simulation task being executed in the target node, the simulation task is determined as a task to be simulated, and the priority of the task to be simulated is reduced to perform the step of allocating the task to be simulated to a corresponding task queue according to the priority of the task to be simulated.

[0160] In some optional implementations, the determination module 702 is further configured to:

[0161] When the node cluster does not include an idle node whose occupation state is an idle state, obtaining the first priority of the simulation task executed in the node cluster;

[0162] The tasks to be simulated with a priority higher than the first priority in the task queue are determined as target simulation tasks.

[0163] In some optional implementations, the acquisition module 701 is further used to:

[0164] Estimate the execution cycle of the task to be simulated;

[0165] The priority of the task to be simulated is set according to the execution cycle, and the task to be simulated is assigned to the corresponding task queue according to the priority.

[0166] In some optional implementations, the acquisition module 701 is further used to:

[0167] Obtaining an execution time interval pre-set for each priority level, wherein the length of the execution time interval is inversely proportional to the height of the priority level;

[0168] The priority of the task to be simulated is set based on the execution time interval to which the execution cycle belongs.

[0169] In some optional embodiments, the device is also used for:

[0170] Based on the preset functional modules, the layout corresponding to the target chip is divided to obtain a sub-layout;

[0171] Based on the design parameters of the sub-layout, the tasks to be simulated corresponding to the sub-layout are generated.

[0172] In some optional embodiments, the device is also used for:

[0173] When the execution result of the task to be simulated does not meet the design parameters, the component parameters in the sub-layout of the task to be simulated are adjusted, and the task to be simulated is allocated to a preset position of the corresponding task queue according to the adjusted priority of the task to be simulated.

[0174] In some optional embodiments, the device is also used for:

[0175] Monitor the waiting time of tasks to be simulated in the task queue;

[0176] When the waiting time exceeds a preset upper limit value corresponding to the priority of the task to be simulated, the priority of the task to be simulated is increased to a second priority;

[0177] Allocate the task to be simulated to the task queue corresponding to the second priority level.

[0178] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0179] The simulation resource scheduling device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0180] The embodiment of the present invention also provides a computer device having the above Figure 7 The simulation resource scheduling device shown.

[0181] See also Figure 8 , Figure 8 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 8 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 8 A processor 10 is taken as an example.

[0182] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0183] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0184] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0185] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0186] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0187] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0188] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of computer program instructions in computer-readable media includes, but is not limited to, source files, executable files, installation package files, etc., and accordingly, the way in which computer program instructions are executed by a computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0189] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A simulation resource scheduling method, characterized in that: The method comprises: Acquire the tasks to be simulated corresponding to the target chip, and assign the tasks to be simulated to the corresponding task queues according to the priorities of the tasks to be simulated; Determine the task to be simulated that satisfies the priority condition in the task queue as the target simulation task; A target node that meets the execution condition is determined in the node cluster, and the target simulation task is allocated to the target node, so as to execute the target simulation task based on the target node, wherein the running state of the target node is an idle state or an occupied state.

2. The method according to claim 1, characterized in that The step of determining a target node that meets a priority condition in a node cluster and allocating the target simulation task to the target node includes: Determine whether the node cluster includes an idle node whose occupation state is an idle state; If so, determining the idle node as a target node, and assigning the target simulation task to the target node; If not, determining in the node cluster that the priority of the simulation task being executed is lower than the occupied nodes of the target simulation task; Acquire preset association information, wherein the preset association information is used to indicate an association relationship between the priority of the simulation task and the corresponding total execution cycle; Based on the preset association information, obtaining a target total execution cycle corresponding to the priority of the simulation task in the occupied node; Determine, from among the occupied nodes, a target node whose simulation task execution time exceeds the target total execution period; The simulation task being executed in the target node is terminated, and the target simulation task is allocated to the target node.

3. The method according to claim 2, characterized in that The method further comprises: After terminating the simulation task being executed in the target node, the simulation task is determined as a task to be simulated, and the priority of the task to be simulated is reduced to perform the step of allocating the task to be simulated to a corresponding task queue according to the priority of the task to be simulated.

4. The method according to claim 1, characterized in that: The step of determining the task to be simulated that satisfies the preset priority in the task queue as the target simulation task comprises: When the node cluster does not include an idle node whose occupation state is an idle state, obtaining a first priority of a simulation task executed in the node cluster; The to-be-simulated task in the task queue whose priority is higher than the first priority is determined as a target simulation task.

5. The method according to claim 1, characterized in that The step of obtaining the task to be simulated corresponding to the target chip and assigning the task to be simulated to a corresponding task queue according to the priority of the task to be simulated includes: estimating the execution cycle of the task to be simulated; The priority of the task to be simulated is set according to the execution cycle, and the task to be simulated is allocated to a corresponding task queue according to the priority.

6. The method according to claim 5, characterized in that The step of setting the priority of the task to be simulated according to the execution cycle includes: Obtaining an execution time interval pre-set for each priority level, wherein the length of the execution time interval is inversely proportional to the height of the priority level; The priority of the task to be simulated is set based on the execution time interval to which the execution cycle belongs.

7. The method according to claim 1, characterized in that The method further comprises: Based on the preset functional modules, the layout corresponding to the target chip is divided to obtain sub-layouts; Based on the design parameters of the sub-layout, a task to be simulated corresponding to the sub-layout is generated.

8. The method according to claim 7, characterized in that The method further comprises: When the execution result of the task to be simulated does not meet the design parameters, the component parameters in the sub-layout of the task to be simulated are adjusted, and the task to be simulated is allocated to a preset position of the corresponding task queue according to the adjusted priority of the task to be simulated.

9. The method according to claim 1, characterized in that: The method further comprises: Monitoring the waiting time of tasks to be simulated in the task queue; When the waiting time exceeds a preset upper limit value corresponding to the priority of the task to be simulated, raising the priority of the task to be simulated to a second priority; Allocate the task to be simulated to a task queue corresponding to the second priority level.

10. A simulation resource scheduling device, characterized in that: The device comprises: An acquisition module, used for acquiring tasks to be simulated corresponding to the target chip, and allocating the tasks to be simulated to corresponding task queues according to the priorities of the tasks to be simulated; A determination module, used for determining the to-be-simulated task that satisfies the priority condition in the task queue as a target simulation task; An allocation module is used to determine a target node that meets the priority condition in the node cluster and allocate the target simulation task to the target node so as to execute the target simulation task based on the target node, wherein the operating state of the target node is an idle state or an occupied state.

11. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the simulation resource scheduling method according to any one of claims 1 to 9 by executing the computer instructions.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the simulation resource scheduling method according to any one of claims 1 to 9.

Citation Information

Cited By

  • Simulation test method, device and system and simulation equipment

    CN119885997A

  • Hardware simulation system, control method, electronic equipment and storage medium

    CN119903791A