A Profit-aware Distributed Data Center Load Management Method

Abstract

The present invention discloses a profit-aware distributed data center load management method, which is based on the architecture of front-end proxy server, global dispatcher and local manager, with the goal of maximizing the operating profit of distributed data center operators, Taking into account the differences in electricity prices at different times and locations, the energy efficiency of data centers, and the quality of service requirements, jointly optimize the load distribution of distributed data centers, the number of active servers, and the operating frequency of CPUs on active servers, so as to achieve high quality of service and Effective compromise between low energy consumption, so that distributed data center operators can obtain more profits.

Description

Technical field [0001] The invention relates to the field of load management of distributed data centers, and in particular to a layered load management method for maximizing the profits of data center operators. Background technique [0002] With the prevalence of online services such as search engines, online games and social networks, data centers, as the core physical platform supporting online services, have begun to enter a stage of rapid development. Many data center operators, including Amazon, Google, Microsoft, Yahoo, and IBM, have newly deployed a large number of data centers around the world to provide computing, storage, and network resources for the growing number of online services. For data center operators, the most concerned issue is how to maximize the operating profit of distributed data centers. [0003] The profit of a data center operator is operating income minus operating costs. Operating costs are mainly determined by the power consumption of distributed ...

AI Technical Summary

Problems solved by technology

[0004] In a distributed data center, the quality of service is positively correlated with energy consumption and service volume. The higher the service quality, the higher the energy consumption required, and the higher the relative service volume. The center uses higher power consumption for calculations, resulting in an increase in the energy consumption cost of the data center. If the quality of service is reduced in order to save energy consumption, the service volume per unit time will be reduced, resulting in a decrease in customer satisfaction, and fines will be paid to customers. Reduced data center revenue

Since the current solutions do not take into account the above-mentioned impacts, it is impossible to effectively compromise between high-quality services and low energy consumption, thereby failing to obtain higher profits

Images