Network storage virtualization scheduling method under private cloud condition

Abstract

The invention relates to the technical field of cloud computing, in particular to a network storage virtualization scheduling method under the private cloud condition. In the scheduling method, a task analyzing module, a resource monitoring and maintaining module, a scheduler module and the like are included; the task analyzing module analyzes the task requirements according to a request of a user, the resource monitoring and maintaining module is responsible for resource allocating, releasing, updating and the like and is also responsible for network state monitoring in physical resources, and the scheduler module conducts task scheduling through a utility computing method according to information of the task analyzing module and the resource monitoring and maintaining module and starts different scheduling strategies according to task resource requesting and resource releasing. According to the method, the different scheduling strategies can be conducted according to the different conditions through the utility computing method, the scheduling complexity is effectively reduced, utility maximization is achieved, the scheduling timeliness is guaranteed, and the demands and earnings of the user are met.

Description

technical field [0001] The invention relates to the technical field of cloud computing, in particular to a network storage virtualization scheduling method under the condition of a private cloud. Background technique [0002] Cloud computing is a new business model. Its core idea is to manage and schedule a large number of resources connected by the network in a unified manner, forming a resource pool to provide users with on-demand services. Cloud computing includes private cloud and public cloud. Among them, the private cloud is built for the sole use of a company or organization, and has attracted attention because it can provide more private data, higher security and better service quality. [0003] The main problem with private clouds is the spike problem. If the private cloud is built according to all the constraints of the workload, the peak load will require the private cloud to invest more hardware resources, resulting in idle and waste of resources. If the constr...

AI Technical Summary

Problems solved by technology

[0004] The current scheduling methods mainly have the following problems: 1) The current scheduling methods are mainly based on public clouds or hybrid clouds, and rarely based on private clouds. The scheduling goals and scheduling space of public clouds and private clouds are different; 2) Most of the scheduling methods All consider the user's QoS, or the utilization rate of resources, etc., and few scheduling methods consider how to meet the needs of users under the condition of network constraints

The task scheduling method of the invention only considers the physical limitations of the bottom layer, but does not consider the real-time requirements of the tasks, which may cause the QoS of the service to be unsatisfactory and the user experience to be poor

Images