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

Method and apparatus for allocating gpus

A technology for allocating requests and request allocation, applied in the field of allocating GPUs, can solve the problem of not considering the connection affinity between GPUs, and achieve the effect of improving utilization.

Active Publication Date: 2021-08-27
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, when the existing technology allocates GPUs for training tasks, it usually does not consider the connection affinity between GPUs, and only ensures that the allocated GPUs are located on the same node.

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 and apparatus for allocating gpus
  • Method and apparatus for allocating gpus
  • Method and apparatus for allocating gpus

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain related inventions, rather than to limit the invention. It should also be noted that, for the convenience of description, only the parts related to the related invention are shown in the drawings.

[0033] It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and embodiments.

[0034] figure 1 An exemplary system architecture 100 is shown to which embodiments of the method for allocating GPUs or the apparatus for allocating GPUs of the present application can be applied.

[0035] Such as figure 1As shown, the system architecture 100 may include t...

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 present application discloses a method and an apparatus for allocating GPUs. A specific implementation of the method includes: receiving a GPU allocation request, wherein the GPU allocation request includes the number N of GPUs requested to be allocated, and N is a positive integer; determining an idle GPU and a first link between an idle GPU, wherein, The first link is used to provide data communication between two GPUs, and the bandwidth of the first link is greater than a preset threshold; based on the idle GPU and the determined first link, a connection topology diagram is generated, wherein the connection The topology diagram includes vertices corresponding to idle GPUs and connection lines corresponding to the first link; based on the generated connection topology diagram, N GPUs are selected from idle GPUs as GPUs requested in the GPU allocation request. This embodiment improves the utilization rate of the GPU.

Description

technical field [0001] The embodiments of the present application relate to the field of computer technologies, and in particular to a method and an apparatus for allocating GPUs. Background technique [0002] In the era of big data, the processing of massive data has higher requirements on the performance of computers. Compared with CPU (Central Processing Unit, central processing unit), GPU (Graphics Processing Unit) has greatly improved computing performance, and has been widely used in training tasks such as floating-point computing, general computing, and image processing. . However, when assigning GPUs to training tasks in the prior art, the connection affinity between GPUs is usually not considered, and only the allocated GPUs are guaranteed to be located on the same node. Contents of the invention [0003] Embodiments of the present application propose a method and an apparatus for allocating GPUs. [0004] In the first aspect, an embodiment of the present appli...

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 Patents(China)
IPC IPC(8): G06F15/163G06T1/20
CPCG06F15/163G06T1/20
Inventor 叶子超
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD