Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Large-scale parallel program property-predication realizing method

A technology of performance prediction and implementation method, which is applied in the direction of concurrent instruction execution, machine execution device, software testing/debugging, etc. It can solve problems such as low precision, inability to calculate time, and low accuracy of parallel program performance prediction, and achieve the effect of less resources

Active Publication Date: 2010-02-17
TSINGHUA UNIV
View PDF0 Cites 26 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, there are some extrapolation methods to obtain the sequential computing time of a strongly scaled parallel program, but the accuracy of the extrapolation methods is relatively low
[0008] In short, the current existing technology cannot obtain the accurate sequential calculation time in parallel programs, especially for large-scale parallel programs, resulting in low performance prediction accuracy of parallel programs

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Large-scale parallel program property-predication realizing method
  • Large-scale parallel program property-predication realizing method
  • Large-scale parallel program property-predication realizing method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0053] The performance prediction method of parallel programs based on representative replay proposed by the present invention requires two hardware platforms: the first is the main platform, which is a general cluster system, and the memory of the main platform must meet the requirements of the parallel program to be predicted. requirements; the second is a child node of the target system. This node is used to collect calculation vectors of representative processes, and the acquisition method will be described in detail below. If the target platform node system is still under development, a hardware emulator for the target platform node is required. The software platform required by the prediction system includes: a public network simulator SIM-MPI and a communication record collection software FACT. The description is as follows in conjunction with the accompanying drawings.

[0054] Such asfigure 1 Shown:

[0055] S1: Collect communication sequences and computation vecto...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a large-scale parallel program property-predication realizing method based on representative replay, belonging to the technical field of parallel program property prediction. The method comprises the following steps: collecting communication sequences and calculation vectors of parallel programs, analyzing calculation similarity of the processes, selecting a representativeprocess, recording the communication content of the representative process, using a calculation node of a target platform to replay the representative process, obtaining the sequence calculation timeof the representative process, and using the calculation time to replace the calculation time of other processes; obtaining the communication records of the parallel programs; and using a network simulator to automatically predict the final program property. By using the invention, only few hardware sources can be used, and the accurate parallel program property prediction can be obtained.

Description

technical field [0001] The invention relates to the technical field of parallel program performance prediction. Background technique [0002] Currently, massively parallel computers include tens of thousands of computing nodes. In the development process, it generally takes several years of development time and high manpower and financial resources. For designers of these computers, being able to predict the performance of an application on a future designed system during the design phase is of great significance for early detection of system bottlenecks and improvement of system design. [0003] Accurately predicting the performance of an application on a future system can not only assist in the design of future computers, but also help application developers discover application bottlenecks before the future system is available, thereby optimizing and improving program performance. [0004] However, accurately predicting the performance of parallel programs on future syst...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F11/36G06F9/38
Inventor 郑纬民陈文光翟季冬
Owner TSINGHUA UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com