Data parallelization processing method, system and device and storage medium

A processing method and technology of a processing system, applied in data parallel processing methods, equipment and storage media, and system fields, can solve problems such as inability to fully utilize CPU and computing card computing resources

Active Publication Date: 2020-05-12
SHENZHEN CORERAIN TECH CO LTD
View PDF6 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Most of the graph reasoning methods used in related technologies are single-threaded operations with asynchronous calculations, and this graph reasoning method is likely to fail to fully utilize the computing resources of the CPU and computing cards.

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
  • Data parallelization processing method, system and device and storage medium
  • Data parallelization processing method, system and device and storage medium
  • Data parallelization processing method, system and device and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] figure 1 It is a flow chart of a data parallel processing method provided by Embodiment 1 of the present invention. This embodiment is applicable to graph reasoning with multiple logical relationships, and the method can be executed by a host. like figure 1 As shown, a data parallel processing method includes S110 to S150.

[0032] S110. Confirm at least three first computing nodes with a logical relationship from a plurality of first computing nodes, and define the at least three first computing nodes with a logical relationship as a first parallel node group, and the first parallel node group includes the first parallel node group A front node and at least two first rear nodes.

[0033] In this embodiment, at least three first computing nodes having a logical relationship mean that the three first computing nodes include at least one first previous node located upstream in the logical relationship and at least two nodes directly connected to the first previous node...

Embodiment 2

[0045] Embodiment 2 of the present invention is an optional embodiment based on Embodiment 1. figure 2 It is a flow chart of a data parallel processing method provided by Embodiment 2 of the present invention. like figure 2 As shown, the data parallel processing method in this embodiment includes S201 to S216.

[0046] S201. Determine whether at least three first computing nodes include a first front node and at least two first rear nodes having a logical relationship.

[0047] In this embodiment, after receiving multiple first computing nodes, select at least three first computing nodes among these first computing nodes and judge whether there is a logical relationship between these three computing nodes, that is, whether there is a first node and at least two first post nodes. Taking the computing node calculation of the neural network as an example, the neural network generally has multiple layers, that is, multiple logically connected computing nodes. , the first pos...

Embodiment 3

[0080] image 3 It is a schematic structural diagram of a data parallel processing system provided by Embodiment 3 of the present invention. like image 3 As shown, the data parallel processing system 300 of this embodiment includes: a screening module 310 , a first acquisition module 320 , a first calculation module 330 , a second acquisition module 340 and a second calculation module 350 .

[0081] A screening module 310, configured to confirm at least three first computing nodes having a logical relationship from a plurality of first computing nodes, defining the at least three first computing nodes having a logical relationship as a first parallel node group, the first parallel The node group includes a first front node and at least two first rear nodes;

[0082] The first obtaining module 320 is used to obtain the first input data model of the first previous node and generate the first input tensor of the first previous node;

[0083] The first calculation module 330 i...

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 data parallelization processing method, system, and device and a storage medium. The method comprises the following steps: defining at least three first computing nodes witha logical relationship as a first parallel node group, the first parallel node group comprising a first front node and at least two first rear nodes; obtaining a first input data model of a first front node and generating a first input tensor of the first front node; calculating a first output tensor of the first front node according to the first input data model and the first input tensor; obtaining second input data models of at least two first back nodes and taking the first output tensor as a second input tensor; and respectively calculating second output tensors of the at least two firstrear nodes according to the second input data model and the second input tensor to obtain a first calculation result of the first parallel node group. According to the method, the technical effect offully utilizing CPU and operation card resources is achieved through overlapping operation of multiple input graph reasoning.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of network topology graph inference, and in particular, to a data parallel processing method, system, device, and storage medium. Background technique [0002] Deep learning networks are usually trained by algorithms. In most cases, algorithm developers tend to use existing public deep learning frameworks for model training, and most public deep learning frameworks are for central processing units / graphics processing units (Central Processing Unit / Graphics Processing Unit, CPU / GPU ) are designed for such computing devices. CPU / GPU adopts traditional instruction set architecture, which has low architecture efficiency and high flexibility. With the development of deep learning related technologies, the requirements for computing power are getting higher and higher. The architectural efficiency defect of the instruction set in the related art can no longer meet the requirements of the a...

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): G06T1/20G06N3/063
CPCG06T1/20G06N3/063G06F9/5077Y02D10/00G06F9/3867G06F15/7821G06F9/4881G06F9/5022G06F9/5066
Inventor 马恺熊超牛昕宇蔡权雄
Owner SHENZHEN CORERAIN TECH CO LTD
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