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

Cluster scheduling system and method

A cluster scheduling and cluster technology, applied in the field of information processing, can solve problems such as inability to perform tasks and low efficiency

Pending Publication Date: 2021-05-04
新疆睿思高科信息科技有限公司
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a cluster scheduling system, which solves the technical problem that the prior art cannot perform sequential execution of tasks according to the execution dependencies between tasks, resulting in low execution efficiency

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
  • Cluster scheduling system and method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] An embodiment of a cluster scheduling system of the present invention is basically as attached figure 1 Shown: includes:

[0033] a determination unit, configured to determine resources and scheduling parameters in the shared cluster;

[0034] A judging unit is used to periodically check whether a checkpoint needs to be set;

[0035] The setting unit is used to determine the tasks that need to be checked when the checkpoint needs to be set, and set the checkpoint for it; specifically, it includes: scheduling a predefined workflow, and the predefined workflow includes multiple groups that exist between Execute tasks that depend on the sequence, among which, a group of tasks without pre-dependencies is the task that needs to be executed first; the tasks are segmented to obtain task data segments, and the task data segments are distributed to the execution terminal for execution, and the task data The number of shards is smaller than the number of execution terminals; wh...

Embodiment 2

[0050] The only difference from Embodiment 1 is that an update unit is further included, which updates the waiting queue of running tasks when a new task arrives or the resource usage status of the shared cluster changes. For the resource usage status of the shared cluster to change, it includes two situations: a task is completed and various available resources of the shared cluster increase. For the former, that is, when a task is completed, the overall required resources are correspondingly reduced, so it is equivalent to the "implicit" increase of available resources; for the latter, when the various available resources of the shared cluster increase, the available resources In both cases, the waiting queue of running tasks should be updated to maximize the utilization of resources as much as possible, thereby improving execution efficiency.

[0051] When the setting unit determines the tasks that need to be set checkpoints, it also includes: obtaining the priority of each...

Embodiment 3

[0053] The only difference from Embodiment 2 is that this system is carried on a desktop computer and is used for the daily work of the operator of the enterprise, and the operator usually sits in front of the desktop computer. When scheduling the workflow, the operator's acceptable maximum waiting time is also considered, and the acceptable maximum waiting time has the highest priority. That is to say, when the waiting time of an operator's task is greater than or equal to the maximum acceptable waiting time, the operator's task will be executed first.

[0054] In this embodiment, the waiting time for a certain operator's tasks can be predicted according to the number, progress and execution speed of the tasks ahead of him, and the details can be performed with reference to the existing technology. However, for the maximum acceptable waiting time, different operators have different values ​​due to their individual factors such as patience and personality. For example, an ope...

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 relates to the technical field of information processing, in particular to a cluster scheduling system which comprises a determining unit used for determining resources and scheduling parameters in a shared cluster; a judging unit which is used for regularly checking whether a checking point needs to be set; a setting unit which is used for determining a task needing to set the checking point when the checking point needs to be set, and setting the check point for the task; and multiple execution units which are used for receiving the task data fragments and executing the task data fragments. When the checking point needs to be set, the task execution sequence is established according to the execution dependency relationship between the tasks, and the technical problem of low execution efficiency in the prior art is solved.

Description

technical field [0001] The invention relates to the technical field of information processing, in particular to a cluster scheduling system and method. Background technique [0002] With the development of modern cloud computing technology, shared clusters make cloud computing platforms a reality. Most commercial cloud platforms use a preemptive task scheduling method to avoid resource shortages in the cluster. When resources in the system are tight, high-priority tasks will be executed first, and the execution process of low-priority tasks will be temporarily interrupted. Restart the execution process of running low priority tasks when sufficient. Although the preemptive task scheduling method can ensure that high-priority tasks are completed as soon as possible, it will also increase the response time of low-priority tasks, consume a large amount of resources, and reduce cloud work efficiency. [0003] In this regard, Chinese patent CN106776395A discloses a task scheduli...

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
CPCG06F9/4881G06F9/5038G06F2209/5017G06F2209/5021G06F2209/484
Inventor 崔永强
Owner 新疆睿思高科信息科技有限公司
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