Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Resource management method and system facing GPU (Graphic Processing Unit) cluster

A GPU cluster and resource management technology, applied in the field of high-performance computing, can solve the problems of heavy load on management nodes and waste of GPU resources, and achieve the effect of reducing the load

Active Publication Date: 2013-10-23
HUAZHONG UNIV OF SCI & TECH
View PDF2 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a GPU cluster-oriented resource management method and system, the purpose of which is to solve the technical problems of heavy management node load and waste of GPU resources existing in the prior art

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
  • Resource management method and system facing GPU (Graphic Processing Unit) cluster
  • Resource management method and system facing GPU (Graphic Processing Unit) cluster
  • Resource management method and system facing GPU (Graphic Processing Unit) cluster

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0047] In order to make the objectives, technical solutions and advantages of the present invention clearer, the following further describes the present invention in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0048] Such as figure 2 As shown, the GPU cluster-oriented resource management method of the present invention includes the following steps:

[0049] (1) The main management node establishes a resource information table and a task information table; specifically, the resource information table records the node number, CPU number, idle CPU number, GPU number, and idle GPU number of each node in the clust...

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 resource management method facing a GPU (Graphic Processing Unit) cluster, which comprises the following steps: a main management node establishes two charts (a resource information chart and a task information chart); the main management node receives a new task; the main management node judges whether the task is a CPU (Central Processing Unit) task or a GPU task; the main management node seeks free resource meeting the requirement of the task; if the task is a CPU task, a secondary management node conducts pretreatment on the data of the task, and dispensing pieces of the data to all nodes managed by the secondary management node for calculation, the main management node reclaims CPU resource related to all the nodes managed by the secondary management node according to the number of the task after calculation; if the task is a GPU task, the main management node reclaims the GPU resource related to all the nodes managed by the secondary management node according to the number of the task when GPU calculation is detected to be finished; meanwhile, the CPUs of all the nodes managed by the secondary management node are used for post-processing of a result, and the post-processing is finished. According to the invention, CPU resource and the GPU resource are treated differently; through the detection of the task, free GPU resource can be reclaimed fast.

Description

Technical field [0001] The present invention belongs to the field of computer high-performance computing, and more specifically, relates to a resource management method and system for GPU clusters. Background technique [0002] In recent years, with the continuous development of high-performance computing, GPU clusters have received more and more attention. The high performance of GPU clusters is mainly due to its massively parallel multi-core structure, high throughput in multi-threaded floating-point arithmetic, and the use of large on-chip caches to significantly reduce the time required for a large amount of data movement. GPU clusters not only make a huge leap in speed and performance, but also significantly reduce the requirements for space, energy, and cooling. [0003] However, the current management of GPU clusters mainly refers to the management mode of CPU clusters, scheduling for CPU cores, and adopts a single centralized unified management mode: the entire cluster has...

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
CPCG06F9/5011
Inventor 金海郑然冯晓文朱磊
Owner HUAZHONG UNIV OF SCI & TECH
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
Eureka Blog
Learn More
PatSnap group products