Method for creating computing power container, computing power platform, electronic equipment and storage medium

Abstract

The invention provides a method for creating a computing power container, a computing power platform, electronic equipment and a storage medium, and relates to the field of big data processing of a cloud technology. The method comprises the steps of obtaining a target task through an affinity task cache; through the affinity task cache, determining a first resource pool corresponding to the target task in N affinity task resource pools corresponding to N time periods according to the time period of the target task, wherein n > 1; and creating a computing power container for the target task in the first resource pool. According to the method, the computing power resource pool can be regularized from the time dimension, the computing power resource pool is prevented from being scattered on the time granularity, and the computing power fragmentization is reduced.

Description

Technical field [0001] The present application relates to the field of cloud technology, in particular to large data processing in the cloud technology, and more particularly to a method, power platform, electronic device, and storage medium for creating a force container. Background technique [0002] Up to now, the industry's technical solution for "reduced integrity fragmentation" is mainly: After the user submits AI training, from the perspective of resources to meet the needs of users, the saturation priority algorithm is usually used, that is, lowers from the perspective of resources. Defragmentation, prioritize fragmentation debris; however, the saturation priority algorithm is valid on the scene of the distribution of resources, and for the resource scenarios of static production delivery, and for the dynamic scene, even if you are full of priority Demand, or a fragmentation. [0003] Therefore, for a diversified resource allocation scenario, it is urgent to create a more...

AI Technical Summary

Problems solved by technology

[0002] So far, the industry's technical solutions for "reducing computing power fragmentation" are mainly: after users submit AI training, to meet user needs from the perspective of resources, a saturation priority algorithm is usually adopted, that is, to reduce computing power from the perspective of resources Fragmentation, giving priority to the distribution of fragmented computing power equipment; however, in the scenario of allocating resources, the saturation priority algorithm is effective for static production and delivery resource scenarios, and for task-oriented dynamic scenarios, even according to the saturation priority packing user's demand, or there will be fragmentation

Images