Optimal task scheduling method and system based on pseudo-tree structure in data center network

Abstract

The invention discloses an optimal task scheduling method and system based on a pseudo-tree structure in a data center network, wherein the mapping cost is the target; a constraint optimization problem model COP is established according to a workload request WK and a substrate network SN; a pseudo-tree is established for a virtual node in the workload request WK; starting from a leaf node of the pseudo-tree, a parent node is reported to step by step and an optimal partial solution set is generated until a root node of the pseudo-tree generates a final optimal partial solution set; starting from the root node of the pseudo-tree, a solution message is transmitted to the child node step by step; and each node confirms the value according to the solution message. The optimal task scheduling method and system based on a pseudo-tree structure in the data center network in the invention can realize applications such as virtual network mapping in a moderately sized environment and resource allocation under a data center network, thereby realizing efficient and even optimal task resource scheduling.

Description

technical field [0001] The invention relates to an optimal task scheduling technology under a data center network, in particular to an optimal task scheduling method and system based on a pseudo-tree structure under a data center network. Background technique [0002] The Task Scheduling Problem (TSP) under the data center network refers to mapping a workload (Workload, WK) request with virtual nodes and virtual link constraints to an infrastructure network (Substrate Network, SN for short). A process in resources, where virtual nodes are mapped to physical nodes in SN, and virtual links are mapped to physical paths in SN, and the resource constraints of virtual nodes and virtual links in WK are satisfied. TSP is an important hotspot in the field of data center network research. It is a typical multi-objective optimization problem. It has attracted extensive attention in the fields of data center virtualization and network virtualization. [0003] In the era of cloud comput...

AI Technical Summary

Problems solved by technology

[0005] The technical problem to be solved by the present invention: Aiming at the above-mentioned problems of the prior art, an optimal task scheduling method and system based on a pseudo-tree structure under a data center network is provided. The present invention models the task scheduling problem as an Constrained optimization problem, and organize the nodes in the workload request into a pseudo-tree structure, and then find the optimal scheduling solution through search and backtracking ideas, to achieve virtual network mapping in moderate-scale environments, resource allocation in data center networks and other applications, So as to achieve efficient and even optimal task resource scheduling

Images