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

Intelligent adjusting and optimizing method for high-performance calculation of user quota

A high-performance computing and user technology, applied in energy-saving computing, computing, resource allocation, etc., can solve the problems of long waiting time for users, waste of energy, and low resource utilization, so as to improve resource utilization and energy efficiency, and ensure fairness. sexual effect

Active Publication Date: 2021-06-25
NORTHWESTERN POLYTECHNICAL UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention aims at problems such as low resource utilization rate, waste of energy, or long waiting time of users that may be caused by high-performance computing user quotas. The present invention proposes an intelligent optimization method for high-performance computing user quotas

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
  • Intelligent adjusting and optimizing method for high-performance calculation of user quota

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] Now in conjunction with embodiment, accompanying drawing, the present invention will be further described:

[0028] The specific steps of the user quota intelligent tuning provided by the present invention are as follows:

[0029] (1) Periodically obtain the current resource utilization and job running status of all users. Specifically, it includes the number of computing nodes and processors that have been used, the number of idle computing nodes and processors, the number of jobs that are running, the status of jobs queued by users, and the status of jobs that users are banned for exceeding quotas;

[0030] (2) If there are currently idle resources and prohibited jobs, query the quotas of the prohibited users, and decide whether to increase the quotas of users with prohibited jobs. If the jobs of multiple users are banned, the dynamic priority of the banned jobs will be calculated, and the user with the highest dynamic priority of the job will be given priority to ad...

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 an intelligent adjusting and optimizing method for high-performance computing user quota, and belongs to the technical field of resource management and job scheduling of high-performance computing clusters. The method comprises the following steps: periodically obtaining a current resource utilization rate and job operation conditions of all users; if the idle CPU core exists and the forbidden operation exists, judging whether the quota of the user with the forbidden operation is increased or not; selecting preferentially adjusted users based on the job dynamic priority; if no idle CPU core exists and queuing operation exists, judging whether the quota value of the user is reduced or not; and based on an intelligent tuning algorithm, modifying the maximum number of jobs, the maximum number of queuing jobs and the maximum number of processors of the selected user, and achieving intelligent increase or decrease of the user quota. According to the method, the resource utilization rate and the energy efficiency of the high-performance computing system can be improved.

Description

technical field [0001] The invention relates to a strategy-based intelligent optimization method for user quotas in a Linux high-performance computing cluster, and belongs to the technical field of resource management and job scheduling of the high-performance computing cluster. Background technique [0002] Small and medium-sized Linux high-performance computing clusters are very popular and are deployed in many universities, research institutes, and laboratories. Due to the large number of users and limited resources, some commonly used resource management and job scheduling software, such as OpenPBS and Maui, will set quotas for high-performance computing users, limiting the maximum number of CPU cores that a user can occupy, the maximum number of jobs that can be run, and the maximum queue. number of jobs, etc. However, the high-performance computing cluster resources used by many users are very dynamic. When the cluster load is light, the jobs of some users cannot be r...

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): G06F9/50
CPCG06F9/5038Y02D10/00
Inventor 侯正雄王云岚谷建华赵天海李秀春
Owner NORTHWESTERN POLYTECHNICAL UNIV