Request scheduling and optimization method for spatial detection in distributed green cloud data center

Abstract

The invention discloses a request scheduling and optimization method for spatial detection in a distributed green cloud data center. The method comprehensively considers changes of factors, such as electric energy price, wind speed, solar radiation strength and field air density generated by a thermal power generation mode, at different geographical locations. Aiming at the requests of a pluralityof applications, the method builds a framework for processing multiple types of application requests under a distributed green cloud data center environment, and accordingly, a non-linear constraintoptimization model of request scheduling of an overall cost of a cloud provider, is built, and a penalty function is designed to convert the non-linear constraint optimization model into an unrestraint optimization model, then a mixed element heuristic optimization algorithm based on simulated annealing and bat algorithms is used for solving the model, and thus request scheduling of spatial detection under the distributed green cloud data center environment is achieved. According to the method provided by the invention, all reached application requests can be scheduled to a plurality of greencloud data centers for executing, so that the overall cost of the cloud provider is minimized and the delay time requirements of all application requests are ensured.

Description

technical field [0001] The invention relates to the technical field of cloud computing resource scheduling. More specifically, it relates to a space-aware request scheduling and optimization method in a distributed green cloud data center. Background technique [0002] In order to enable global users to flexibly use resource services, the infrastructure in cloud computing usually includes millions of servers and cooling facilities. The energy cost of green cloud providers (such as Microsoft, Amazon) has increased significantly in recent years as the types and numbers of applications have increased. In 2014, the energy consumption of all data centers in the United States reached 70 billion kWh, which accounted for 2% of the total energy consumption in the United States. Data show that this proportion will reach 4% in 2020. In the United States, more than 57% of energy is obtained by burning fossil fuels such as coal to generate electricity. However, this will bring irreve...

AI Technical Summary

Problems solved by technology

However, the arrival of user application requests has no specific rules, and a single green cloud data center has limited resources, so it cannot handle all application requests.

Images