A method, device and computer storage medium for analyzing gpu performance

Abstract

The embodiment of the present invention discloses a method, device, and computer storage medium for analyzing GPU performance; the method may include: obtaining the instruction list obtained by running the target program in a set environment, the number of threads to be started, and the effect of each thread on all The execution result of each instruction in the above instruction list; Start the thread simulator in the GPU performance model to be analyzed according to the number of threads to be started by the simulation scheduler in the GPU performance model to be analyzed; each thread simulator Each instruction in the instruction list is traversed, and the instruction is executed according to the instruction execution control value of each instruction during the traversal process, so as to measure the duration of executing the traversed instruction; when all the instructions in the instruction list have been traversed , to obtain the total execution time of all the thread emulators executing all the instructions in the instruction list.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of graphics processing units (GPUs, Graphics Processing Units), and in particular, to a method, an apparatus, and a computer storage medium for analyzing GPU performance. Background technique [0002] In GPU performance statistics, Instructions Per Cycle (IPC, Instructions Per Cycle) is a relatively important GPU performance indicator, which represents how many instructions the GPU can process in each clock cycle; The execution time and the main frequency of the system are calculated to obtain the IPC. [0003] In the process of GPU performance statistics, it is usually necessary to model GPU performance. Specifically, the performance of GPU is usually modeled in two ways: one is simulation modeling, such as using software simulation to build a simulation model of GPU, and performing real execution process according to the simulation model to obtain real performance data of GPU; The s...

AI Technical Summary

Problems solved by technology

[0004] For the simulation modeling method, although it can truly simulate the hardware execution process and obtain real simulation data; however, since the simulation model needs to simulate the execution of the real GPU, the running efficiency is low. If the GPU architecture needs to be adjusted, then It is necessary to rebuild the simulation model for the GPU after the architecture adjustment. Therefore, using the simulation modeling method for GPU performance statistics has the disadvantages of relatively poor scalability and long development cycle.

For the analysis and modeling method, the analysis model does not need to simulate the actual operation process of the instruction, but only needs to perform modeling analysis operations on the input instruction information to obtain the performance result data, so the analysis and modeling method is used to calculate the operating efficiency of GPU performance statistics Very high, simple structure design, and strong scalability; however, in the actual implementation process, if the analysis model does not process the output instructions finely, it will cause a large error rate in the final GPU performance statistics

Images