Resource pool allocation method and resource scheduler

Abstract

The present invention discloses a resource pool allocation method and a resource scheduler. The method comprises: dividing an application task into a plurality of subtasks; grouping the plurality of subtasks of the application task into one or more sets according to dependency among the subtasks; establishing an objective function, wherein the objective function is a sum of computational capabilities of occupied resource pools and a sum of bandwidth resources when each of the set uses allocated resource pools to perform a task; for each set of an application task, selecting a corresponding resource pool from a plurality of candidate resource pools as a resource pool allocation scheme of the application task; using a particle swarm algorithm to solve a resource pool allocation scheme which minimizes a value of the objective function; and performing allocation according to the obtained resource pool allocation scheme. According to the resource pool allocation method and the resource scheduler disclosed by the present invention, the objective function is established, and the particle swarm algorithm is used to optimize allocation and management of heterogeneous resource pools in a cloud environment, thereby selecting an optimal resource pool allocation scheme for a cloud computing application task quickly and efficiently.

Description

technical field [0001] The invention relates to the communication field, in particular to a resource pool allocation method and a resource scheduler. Background technique [0002] In the current cloud computing environment, business is usually only completed in a certain cloud computing resource pool. Compared with the traditional serial transaction processing mode, a heterogeneous cloud computing Infrastructure as a Service (IaaS for short) resource pool is proposed. The heterogeneous private cloud and public cloud are combined as a large resource pool for the cloud computing resource scheduler to schedule. In a heterogeneous cloud computing IaaS resource pool, applications and services can be decomposed into smaller components called subtasks. like figure 1 As shown in Fig. 1, an application is divided into 8 subtasks, and the logical sequence of the subtasks of an application is called a workflow, which is usually represented by a directed acyclic graph. A node represe...

AI Technical Summary

Problems solved by technology

Since multiple subtasks are interdependent, and each IaaS resource pool has different computing capabilities and bandwidth capabilities, there are many alternative allocation schemes for selecting different resource pools for each subtask. Solutions may bring different performance / processing capabilities, and there is currently a lack of a solution for selecting the best combination of resource pools for an application

Images