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

Task scheduling method, computer equipment and storage medium

A task scheduling and task technology, applied in computing, program control design, program startup/switching, etc., can solve problems such as no longer applicable application scenarios, achieve the effect of improving overall operating efficiency and reducing the possibility of resource conflicts

Active Publication Date: 2021-02-05
NANJING TRANSWARP INTELLIGENCE CO LTD
View PDF4 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Nowadays, the resources that need to be considered in large-scale heterogeneous parallel clusters are no longer limited to CPU, disk, memory, etc., and the resources of different cluster machines are also quite different. Currently, only The task scheduling method using general simple constraints is no longer suitable for today's complex application scenarios

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
  • Task scheduling method, computer equipment and storage medium
  • Task scheduling method, computer equipment and storage medium
  • Task scheduling method, computer equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] figure 1 It is a flow chart of a task scheduling method provided by Embodiment 1 of the present invention. This embodiment is applicable to task scheduling in large-scale heterogeneous parallel clusters, for example, it can be applied to scenarios such as big data, artificial intelligence, and cloud computing. In large-scale heterogeneous parallel cluster task scheduling in , the method can be executed by a task scheduling device, which can be implemented by software and / or hardware, and can generally be integrated in cluster computer equipment.

[0047] Such as figure 1 As shown, the task scheduling method provided in this embodiment includes the following operations:

[0048] Step 110, acquiring tasks to be scheduled, and determining target resource requirements of the tasks to be scheduled.

[0049] Tasks to be scheduled refer to tasks that require resource scheduling. The task to be scheduled is put into the task queue to be executed, and the system takes out the...

Embodiment 2

[0072] image 3 It is a flow chart of a task scheduling method provided by Embodiment 2 of the present invention. The embodiment of the present invention can be combined with each optional solution in the above-mentioned one or more embodiments. In this embodiment, according to the preset evaluation model, respectively Evaluate each resource allocation candidate subgraph; use the resource allocation candidate subgraph with the best evaluation result as the resource allocation target subgraph.

[0073] Such as image 3 As shown, the task scheduling method provided in this embodiment includes the following operations:

[0074] Step 210, acquiring tasks to be scheduled, and determining target resource requirements of the tasks to be scheduled.

[0075] Step 220, traverse the available resource attribute graph at the current moment, and generate a preset number of resource allocation candidate subgraphs that meet the target resource requirements; wherein, the attributes of the v...

Embodiment 3

[0091] Figure 9 It is a flow chart of a task scheduling method provided by the third embodiment of the present invention. The embodiment of the present invention can be combined with each optional solution in one or more of the above embodiments. In this embodiment, when traversing the available resources at the current moment Before the attribute graph generates at least two resource allocation candidate subgraphs that meet the target resource requirements, it also includes: if it is determined that the current available resources cannot meet the target resource requirements, release the resources occupied by the executing tasks according to the preset resource preemption strategy, and update A property map of available resources at the current moment.

[0092] Such as Figure 9 As shown, the task scheduling method provided in this embodiment includes the following operations:

[0093] Step 310, acquiring tasks to be scheduled, and determining target resource requirements ...

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 embodiment of the invention discloses a task scheduling method, computer equipment and a storage medium. The method comprises the steps of obtaining a to-be-scheduled task, and determining a target resource demand of the to-be-scheduled task; traversing the available resource attribute graph at the current moment, and generating at least two resource allocation candidate sub-graphs meeting thetarget resource demand; wherein attributes of vertexes in the available resource attribute graph and the resource allocation candidate sub-graph represent resource quantity information, and attributes of edges represent resource association information; and determining a resource allocation target sub-graph in the at least two resource allocation candidate sub-graphs, and deploying the to-be-scheduled task according to the resource allocation target sub-graph. The task scheduling scheme is suitable for a large-scale heterogeneous parallel cluster, can perform finer scheduling management on cluster resources, and improves the overall operation efficiency of the cluster.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of task scheduling, and in particular, to a task scheduling method, computer equipment, and storage media. Background technique [0002] With the popularization of big data technology, the scale of modern clusters is getting larger and larger, and there are more and more types of tasks running in the clusters, and the running cycles are getting longer and longer. During the entire life cycle of the cluster, the hardware of the machines in the cluster may be different and not homogeneous, and it has become a trend to run mixed loads on the same cluster. [0003] Generally speaking, task scheduling refers to resource allocation for tasks. In traditional application scenarios, cluster machines and running tasks are single isomorphic, and the scheduling granularity is relatively coarse. Usually, host resources are divided into one or more groups based on memory or CPU (Central Processing U...

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): G06F9/48G06F9/50
CPCG06F9/4881G06F9/5027Y02D10/00
Inventor 唐剑飞
Owner NANJING TRANSWARP INTELLIGENCE CO LTD
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