Data center self-adapting energy consumption management method based on heat load balancing

Abstract

The invention discloses a data center self-adapting energy consumption management method based on heat load balancing. In each sampling time, the following steps are adopted to perform energy consumption management: 1) using an RLS model to estimate server power consumption; 2) optimizing and solving the cost function within constraint conditions to obtain the optimal Up (k); 3) selecting the first group U (k|k) as a real manipulated variable U (k); 4) calculating the number mj of required active servers; 5) uniformly distributing the tasks to each rack. The data center self-adapting energy consumption management method can dynamically distribute server tasks to reduce the total power consumption of the server in the data center and a cooling system.

Description

technical field

[0001] The invention relates to energy consumption management of a network data center, and more specifically relates to an adaptive energy consumption management method of a data center based on thermal load balancing. Background technique

[0002] In recent years, with the vigorous development of cloud computing. The network data center, as its infrastructure, carries a large number of computing, storage and application tasks, and consumes a lot of energy to run and cool the equipment. With the rapid development of cloud computing services, the energy consumption in data centers is increasing.

[0003] Servers in data centers, where user task requests are processed and stored, use a lot of energy and emit a lot of energy. However, due to the limitation of the operating temperature of the electronic equipment, cooling equipment is required to cool the large amount of heat energy generated by the servers in the data center.

[0004] Currently, academia and...

Images