A parallel pattern search space construction system and method

A pattern search and pattern technology, applied in the field of data processing, to achieve the effect of improving data processing efficiency and reducing the requirements for rich experience

Active Publication Date: 2022-03-25
ZHEJIANG LAB
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As the amount of data increases, this data transmission overhead between different computing devices will cause a huge computational burden for distributed data processing systems

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
  • A parallel pattern search space construction system and method
  • A parallel pattern search space construction system and method
  • A parallel pattern search space construction system and method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] In order to make the object, technical solution and technical effect of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0034] Such as figure 1 As shown, a parallel pattern search space construction system of the present invention includes:

[0035] The initializer S101 is used to initialize the input logical tensor and generate the initial logical node topology diagram for the distributed data processing system;

[0036] The iterator S201, including an input tensor parallel mode iterator S202 and an output tensor parallel mode iterator S203, is used to iterate each candidate parallel mode in the parallel mode search space.

[0037] The parallel pattern search space size is related to the unrestricted distribution descriptor, and each candidate parallel pattern in the parallel pattern search space specifies the distribution descriptor of each input tensor of ...

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 present invention provides a system and method for constructing a parallel pattern search space. The method includes: initializing the input logical tensor; calculating and outputting the true value of the logical tensor; constructing all candidate parallel patterns, and determining the iteration content of the input and output tensor parallel patterns; combining Input the tensor parallel mode iteration content, split the input logical tensor into physical tensor; judge the legality of the physical tensor calculation, if legal, calculate the physical tensor result and output it; if illegal, continue to input the tensor parallel mode iteration process; Combining the iteration content of the output tensor parallel mode, merge the physical tensor operation result into a logical tensor; compare the logical tensor truth value with the logical tensor, if they are equal, add the parallel mode to the legal parallel mode search space, and directly continue to input the tensor If they are not equal, continue the iterative process of the output tensor parallel mode until the iteration is completed, repeat the above steps until the input tensor parallel mode iterative process is completed.

Description

technical field [0001] The invention relates to the technical field of data processing, in particular to a parallel pattern search space construction system and method. Background technique [0002] With the popularization of distributed computing, large-scale jobs or large-scale logical tensors will deploy different parts of data to various computing devices of different distributed data processing systems for processing by splitting, and intermediate parameters will be performed during each calculation process. interaction, which will incur data transfer overhead between computing devices. As the amount of data increases, the data transmission overhead between different computing devices will cause a huge computational burden for the distributed data processing system. [0003] In the invention application with the application number "202010090335.2", the announcement number CN110955734B, and the invention title "Distributed Signature Decision-Making System and Method for...

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 Patents(China)
IPC IPC(8): G06F16/953G06F21/64H04L9/32
CPCG06F16/953G06F21/64H04L9/3247
Inventor 杨非陈岱渊华炜鲍虎军钟昊文
Owner ZHEJIANG LAB
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap