Method for hpl test optimization based on memory affinity

An optimization method and an affinity technology, applied in the computer field, can solve problems such as reducing system memory access performance, achieve the effect of improving memory access performance, good promotion and application value, and improving hpl test results

Inactive Publication Date: 2018-03-23
ZHENGZHOU YUNHAI INFORMATION TECH CO LTD
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the auto_numabalaning mechanism will have a certain impact on the operation of the process, because there will be a lot of overhead in the process of process and memory migration, and the overhead consumed in the process of process and memory migration will greatly reduce the memory access performance of the system. further improvement

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 for hpl test optimization based on memory affinity
  • Method for hpl test optimization based on memory affinity

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0030] Such as figure 1 As shown, the hpl test optimization method based on memory affinity of the present invention specifically comprises the following steps:

[0031] S1: Automatically collect the CPU configuration of the test environment through system files.

[0032] The CPU configuration of the test environment is automatically collected through the profs file system of the Linux system. The CPU configuration includes the number of nodes and the number of cores.

[0033] S2: According to the CPU configuration, set the number of parallel processes used during the test run, and the number of parallel processes is equal to the number of cores configured by the CPU.

[0034] S3: Load the kernel module that sets the process affinity, set the scheduling mask of the process through the memory affinity setting algorithm, bind the process to a node, and set the process to apply for memory locally, and finally bind the hpl process on average in each node of the system.

[0035]...

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 method for hpl test optimization based on memory affinity and belongs to the technical field of computers. The method for hpl test optimization based on memory affinity specifically includes the following steps of S1, automatically collecting CPU configuration of a test environment by use of system files; S2, setting the number of parallel processes used during the test run according to the CPU configuration; and S3, loading a kernel module which sets process affinity. The method for hpl test optimization based on memory affinity can avoid the overhead of process or memory migration to greatly improve the system's memory performance and has a good popularization and application value.

Description

technical field [0001] The invention relates to the technical field of computers, and specifically provides an hpl test optimization method based on memory affinity. Background technique [0002] Computers have the advantages of a large amount of stored information and convenient and quick access to information by users, and are widely used. In the SMP (Symmetrical Multi-Processing) hpl (high performancecomputing linpack benchmark) high-performance computing linpack benchmark test in a multi-symmetrical processor environment, the matrix size of the test will be set according to the memory configuration of the test environment, while in the SMP environment cross-node The performance of accessing memory is lower than that of accessing local memory, so cross-node memory access will affect the hpl test. At present, during the hpl test, the system's auto_numabalaning mechanism can be used to dynamically migrate the process and memory during the test process so that the process c...

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): G06F11/36G06F9/50
CPCG06F9/5016G06F11/3664
Inventor 刘长生
Owner ZHENGZHOU YUNHAI INFORMATION 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