Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and device for task scheduling

A task scheduling and scheduling scheme technology, applied in the field of big data processing, can solve problems such as no solutions are given, and achieve the effect of improving throughput, reducing task execution time, and meeting data locality requirements

Active Publication Date: 2018-06-01
CHINA UNITED NETWORK COMM GRP CO LTD
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the industry has not yet given a suitable solution

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
  • Method and device for task scheduling
  • Method and device for task scheduling
  • Method and device for task scheduling

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] Embodiment 1. A task scheduling method, such as figure 1 shown, including:

[0042] S101. When it is judged that the tasks of the first node need to be scheduled to the second node, acquire the current attribute data of each node; the second node is an idle node;

[0043] S102. Determine the combined scheduling scheme with the smallest expected completion time according to the attribute data; the combined scheduling scheme refers to migrating the tasks executed by each node on the directional path to the next node for re-execution according to the corresponding directional path The directional paths corresponding to each of the combined scheduling schemes are different; each of the directional paths includes two or more nodes, the starting point is the first node, and the end point is the second node;

[0044] S103. Migrate the tasks of each node on the corresponding directional path according to the determined combined scheduling scheme.

[0045] In this embodiment, ...

Embodiment 2

[0098] Embodiment 2. A task scheduling device, comprising:

[0099] The update module is used to obtain the current attribute data of each node in the system when it is judged that the task of the first node needs to be scheduled to the second node; the second node is an idle node;

[0100] A planning module, configured to determine a combined scheduling scheme with the smallest expected completion time according to the attribute data; the combined scheduling scheme refers to migrating the tasks performed by each node on the directional path to the next node according to the corresponding directional path The directional paths corresponding to each of the combined scheduling schemes are different; each of the directional paths includes two or more nodes, the starting point is the first node, and the end point is the second node ;

[0101] The migration module is configured to migrate the tasks of each node on the corresponding directional path according to the determined comb...

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 task scheduling method and device; the method includes: when it is judged that the task of the first node needs to be scheduled to the second node, acquiring the current attribute data of each node in the system; the second node is idle node; determine the combined scheduling scheme with the smallest expected completion time according to the attribute data; the combined scheduling scheme refers to migrating the tasks performed by each node on the directional path to the next node for re-execution according to the corresponding directional path The directional paths corresponding to each of the combined scheduling schemes are different; each of the directional paths includes two or more nodes, the starting point is the first node, and the end point is the second node; The determined combined scheduling scheme migrates the tasks of each node on the corresponding directed path. The present invention can meet the requirement of data locality when scheduling and forecasting execution tasks, thereby reducing the execution time of big data processing jobs.

Description

technical field [0001] The invention relates to the field of big data processing, in particular to a task scheduling method and device. Background technique [0002] In the job scheduling problem of a big data processing system, as the data set increases, the cost of moving the data is far greater than the cost of "moving" the data processing module. Therefore, in the big data processing environment, it is necessary to push the calculation to the data instead Not pushing data to computing, that is, obtaining better data locality (DL: data locallity) is the core of obtaining high efficiency of big data processing systems. [0003] At the same time, in the process of big data processing, if a node executes tasks significantly slower than other nodes, the node is marked as a straggler, and the tasks executed on it are marked as straggler tasks. Tasks will greatly prolong the execution time of big data processing batch tasks such as MapReduce, which is defined by Google enginee...

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): G06F9/50G06F9/38
Inventor 雷磊王志军房秉毅
Owner CHINA UNITED NETWORK COMM GRP CO LTD
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com