A method for creating flexible allocation of multi-specification vGPU virtual machines based on OpenStack platform

By enabling multi-special vGPU types on the OpenStack platform and setting trait properties, the problem of only one vGPU graphics card type in the existing technology is solved, and flexible allocation of video memory size and the creation of multi-special vGPU virtual machines are realized, which meets the diversity of graphics card performance needs of different users.

CN113672344BActive Publication Date: 2025-03-21SHANDONG LANGCHAO YUNTOU INFORMATION TECH CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202110894718.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-08-05
Publication Date
2025-03-21
Estimated Expiration
2041-08-05

AI Technical Summary

Technical Problem

On the OpenStack platform, the existing technology can only configure one vGPU graphics card type, which cannot meet the diversity of graphics card performance needs of different users, resulting in reduced graphics card flexibility and cannot meet the needs of multiple scenarios.

Method used

By enabling multi-special vGPU types on the OpenStack platform, binding the corresponding GPU devices, setting the trait properties of the vGPU provider, creating a custom flavor and setting the trait properties, matching the corresponding vGPU type provider through the trait attribute scheduling, achieving flexible allocation of video memory size.

Benefits of technology

It realizes the ability to create multi-spec vGPU virtual machines on the OpenStack platform, meets the diversity of graphics card performance needs of different users, and improves the rational creation of virtual machines and the flexible allocation of graphics card resources.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention provides a method for creating a multi-specification vGPU virtual machine based on the OpenStack platform and flexibly allocating the vGPU virtual machine, which belongs to the field of OpenStack cloud computing technology, including: enabling the vGPU type on the vGPU computing node and binding the corresponding GPU device; setting the trait attribute of the vGPU provider through the openstack cli command; one or a group of trait attributes uniquely identifying a vGPU type; creating the flavor used by the vGPU virtual machine. Automatically allocate video memory to meet the use of graphics cards in different scenarios.
Need to check novelty before this filing date? Find Prior Art