Energy saving method and system for cloud compute data center

Abstract

The invention discloses an energy saving method and an energy saving system for a cloud compute data center, which relate to the technical field of the Internet. The method comprises the steps of: A1, acquiring the quantity of virtual machines of the current data center; S2, sequencing the virtual machines according to the quantity of the virtual machines on each physical machine; S3, judging the minimal lower limit quantity n of needed physical machines; S4, selecting n physical machines which have the most virtual machines as a target unit; S5, using the virtual machines on the physical machines except the physical machines in the target unit as a transfer unit; S6, letting the selected current virtual machine correspond to the target unit, and orderly implementing transfer of the virtual machines according to the correspondence relationship of the transfer unit and the target unit. The energy saving method and the energy saving system for the cloud compute data center perform transfer judgment in sequence from less to more according to the quantity of the physical machines, can transfer the physical machines completely as soon as possible, and implement reduction of the energy consumption without reducing the work performance of the cloud compute data center when the virtual machines are transferred.

Description

technical field [0001] The invention relates to the technical field of the Internet, in particular to an energy-saving method and system for a cloud computing data center. Background technique [0002] Cloud computing is a computing model based on the Internet. It distributes computing tasks on a resource pool composed of a large number of computing nodes, enabling users to obtain computing power, storage space and information services on demand. The processing methods of cloud computing mainly include parallel computing and Computing resource virtualization, etc. The core concept of cloud computing is a resource pool, which is usually a virtualized computing resource that can be maintained and managed by itself, including computing servers, storage servers, and bandwidth resources. [0003] The role of parallel computing is to split large-scale computing tasks, and then dispatch them to nodes in the cloud for distributed parallel computing, and finally collect and organize...

AI Technical Summary

Problems solved by technology

[0006] After the cloud computing data center has been working for a period of time, the active virtual machines will be scattered on different physical machines. When the workload is relatively stable, the number of virtual machines running on some physical machines is small, and these physical machines are at low utilization. state, the baseline energy consumption should still be maintained, resulting in waste of energy consumption. In order to avoid waste of energy consumption in the prior art, it is necessary to aggregate the virtual machines on different physical servers through migration at this time, thereby shutting down or dormant and evacuated physical servers

[0007] The current research on virtual machine migration mainly starts from the perspective of load balancing and considers the application correlation between virtual machines. The current research usually abstracts the virtual machine placement problem into an NP problem, and uses a heuristic method to determine the virtual machine placement. location, the most typical method is to abstract it into a box-packing problem, and only use the box-packing algorithm to migrate it based on the number of virtual machines, which will increase the communication cost of the entire data center and affect the work of cloud computing performance, and can not effectively reduce energy consumption

Images