Resource scheduling method, device, equipment and system

Abstract

The embodiment of the invention provides a resource scheduling method, device, equipment and system, and the method comprises the steps: enabling a first cluster instance to send a resource borrowingrequest to a second cluster instance if the own available resource of the first cluster instance is lower than a threshold value; receiving borrowable resource information returned by the second cluster instance, wherein the borrowable resource information represents borrowable data server resources in the second cluster instance; and the first cluster instance obtaining borrowable data server resources for data processing, so that resource reuse among the cluster instances is realized, resources of each cluster instance are fully used, and the resource utilization rate of the cluster is improved.

Description

technical field [0001] The present invention relates to the technical field of the Internet, in particular to a resource scheduling method, device, equipment and system. Background technique [0002] In today's era of information explosion, the amount of data generated by various network services has grown exponentially. In order to meet the needs of massive data storage capacity, data backup, and data security, distributed file system clusters have emerged as the times require. , simply put, a distributed file system cluster consists of a group of servers that can provide complete business computing logic. [0003] In an application scenario, the distributed file system cluster may be shared by multiple different businesses, or by multiple different tasks of a business or called sub-services. At this time, the distributed file system cluster can be divided into There are multiple cluster instances (or also called unit clusters, sub-clusters, etc.), and each cluster instanc...

AI Technical Summary

Problems solved by technology

However, more and more businesses have large data volume fluctuations. For example, a certain business may only have a short-term sudden increase in data volume. The real-time nature of capacity expansion cannot meet the real-time processing requirements of the current sudden increase in data volume. On the other hand, capacity expansion to cope with sudden data volume growth often leads to a great waste of resources.

Images