Python environment parallel computing method and device based on domestic CPU and OS

Abstract

The invention discloses a python environment parallel computing method and device based on a domestic CPU and an OS, and belongs to the technical field of python language programming. According to the method, a C / C + + thread is used for running multiple cores, python and C / C + + call each other, a part with dense operand is put into C / C + + to be executed, and then an operation result is transmitted to the python; the realization of the method comprises a communication module, a parallel module and a calculation module, wherein the communication module is used for inputting to-be-called python algorithm model data and outputting calculation result data of the calculation module; and the calculation module provides a calculation data type for the communication module and provides a calculation function for the parallel module. The method can solve the problem that the python program is low in operation efficiency when the model is complex and the calculation amount is large, and realizes the utilization of multiple cores by multiple threads.

Description

technical field [0001] The invention relates to the technical field of python language programming, in particular to a python environment parallel computing method and device based on domestic CPU and OS. Background technique [0002] As a powerful scripting language, python can be applied when the model is simple and the calculation is not heavy. However, when the model is more complex and the amount of calculation is large, the short board of python will appear, that is, the calculation speed is slow, and the reason is the problem of GIL lock. [0003] When Python was released in 1989, because the computer had not yet reached the main frequency of 1G, the programs were all running on a single-core computer. It was not until 2005 that multi-core processors were developed, but the concept of threads and processes has long been Appeared. So for Python, the concept of multi-threading and multi-process needs to be supported first. For the process, it is very simple, just open...

AI Technical Summary

Problems solved by technology

[0004] The technical task of the present invention is to provide a python environment parallel computing method and device based on domestic CPU and OS, which can solve the problem of low operating efficiency of python programs when the model is complex and the amount of calculation is large, and realize the utilization of multi-threading to multi-core

Images