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

Processor scale selection method based on parallel computing time shortest estimation model

A parallel computing and estimating model technology, applied in computing, electrical digital data processing, instruments, etc., can solve the problems of shortening parallel software development and application process, long test time, etc. , the effect of speeding up the application process

Pending Publication Date: 2021-10-01
上海行健职业学院
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the above existing problems, the present invention provides a processor scale selection method based on the shortest estimation model of parallel computing time, which can solve the bottleneck problem of long testing time in the development cycle, and greatly shorten the process of parallel software development and application

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
  • Processor scale selection method based on parallel computing time shortest estimation model
  • Processor scale selection method based on parallel computing time shortest estimation model
  • Processor scale selection method based on parallel computing time shortest estimation model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0071] 1. Experimental environment

[0072] Use the Shanghai University supercomputer "Ziqiang 4000" to conduct simulation tests.

[0073] 2. Performance analysis and comparative effect

[0074] In order to fully verify the calculation model proposed in this paper, this experiment ignores t in formula (4) 0 With the influence of o, only the running time of the program and the amount of data corresponding to different parallel strategies are counted.

[0075] The performance analysis process of the data in this experiment includes:

[0076] Step 1: Select different numbers of processor scales, run four types of parallel programs respectively, and record the corresponding running time.

[0077] Step 2: For the four types of parallel programs, give their corresponding time estimation functions, calculate their derivatives, and calculate the processor scale when the derivative is zero.

[0078] Step 3: Compare the experimental running results with the calculation results, and ...

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 processor scale selection method based on a parallel computing time shortest estimation model. The method comprises the steps: calculating the parallel computing time of a given parallel program running in a parallel machine system by building the parallel computing time shortest estimation model, and determining a processor scale reference range when the calculation time is shortest in the parallel machine system. The number of processors of the parallel machine system is adjusted according to an estimated reference value, and the operation efficiency of the parallel machine system can be exerted as much as possible. According to the method, the program running times can be greatly reduced, and the time consumption of parallel program design, development and application is fundamentally saved, so that the application process of parallel software is accelerated.

Description

technical field [0001] The invention relates to the technical field of parallel computing, in particular to a processor scale selection method based on the shortest estimation model of parallel computing time. Background technique [0002] At present, parallel computing occupies an important position in various fields, such as artificial intelligence and big data, which are widely valued today, cannot do without the support of high-performance computing. In fact, the design and development of parallel programs is a relatively complicated process, and it takes a relatively long period before it can be put into application. One of the key factors is the lack of effective estimation methods for parallel program performance testing and processor size, and it is necessary to continuously run tests to obtain corresponding data. As we all know, only large-scale applications may need to use parallel computing for processing. Such software takes a long time to run once. During the d...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/50
CPCG06F9/5044
Inventor 胡悦闫振
Owner 上海行健职业学院
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