Parallel adjusting and performance analyzing method of supporting multi-language multi-platform under isomerized environment

Abstract

The invention supplies multi-language multiplatform parallel debugging and performance analysis method under isomerization environment. The parallel debugging and performance analysis system is divided into monitoring end, concentrating watching process, and long distance running node. It realizes visualization performance analysis and debugging which includes the steps that front end code analysis, auto pile pitching, debugging support and performance analysis, and visualization user management. It supports isomerization parallel environment, multi-language and parallel platform of which architecture irrespective with the language and parallel platform has pile pitching module independent of the concrete programming language and parallel library. It supplies a new method and means for multi-language multiplatform parallel debugging and performance analysis under isomerization environment.

Description

technical field [0001] The invention relates to a parallel debugging and performance analysis method, belonging to the technical field of debugging and monitoring in high-performance parallel computing. Background technique [0002] At present, the research on the network high-performance computing environment has been widely valued, and the construction ideas and implementation forms are also different. According to the form of construction, they can be divided into two categories related to technology: [0003] (1) Cluster computing. Use a high-speed network to connect a group of workstations or microcomputers to form a cluster, or find a group of idle processors on the general network to form a dynamic virtual cluster, so that it can provide high-performance computing services with high cost performance under the control of intermediate management. [0004] (2) Grid computing. Its goal is to organize and manage computing resources, data sources and other devices in a w...

AI Technical Summary

Problems solved by technology

The resulting complexity in the high-performance computing environment, the diversity of parallel programming languages ​​and parallel platforms, have caused greater difficulties for the development of high-quality parallel software, and have also raised new challenges for the debugging and performance monitoring and analysis of parallel programs. higher requirement

[0006] According to the literature search, traditional parallel program debugging and performance analysis tools, such as Dawning 2000 parallel debugger DCDB and ParaVision of the Institute of Computing Technology, Chinese Academy of Sciences, etc., are relatively weak in terms of scalability for different platforms

As for the heterogeneity characteristics in the cluster or grid computing environment, there are no mature related tools in China at present.

[0007] Among the similar foreign tools, except for a few such as TotalView of Etnus, which are provided by relatively good third-party manufacturers, the others are generally provided by parallel computer manufacturers such as SUN, IBM, and HP for their specific models of parallel computers. Therefore, these debugging tools are not compatible with the platform, and some tools only provide parallel program debugging, and cannot perform performance analysis well.

Images