Multi-energy online control method and system for data center under dynamic cloud service request

Abstract

The invention discloses a multi-energy online control system for data centers under dynamic cloud service requests. The multi-energy online control system for data centers under dynamic cloud service requests comprises a system status monitoring module, a load scheduling module and a multi-source energy supply system management module, wherein the load scheduling module comprises a delay-sensitive request scheduling sub-module and a delay tolerant job scheduling sub-module. The system status monitoring module is used for receiving service requests from a user's cloud service every once in a while, determining whether the cloud service request is a delay-sensitive request or a delay tolerant job, and sending the cloud service request belonging to the delay-sensitive request to the delay-sensitive request scheduling sub-module of the load scheduling module while sending the cloud service request belonging to the delay tolerant job to the delay tolerant job scheduling sub-module of the load scheduling module. The multi-energy online control system for data centers under dynamic cloud service requests has the advantages that long-term operating cost of data center energy supply system can be optimized, no system data is needed to be obtained in advance, and no steady state distribution is needed to be assumed.

Description

technical field [0001] The invention belongs to the technical field of data center energy system control, energy storage equipment management and load scheduling, and more specifically relates to an online control method and system for multiple energy sources in a data center under dynamic cloud service requests. Background technique [0002] At present, cloud service providers face three major problems when operating their data centers: (1) Soaring energy consumption and electricity costs, such as Google (annual energy consumption > 1120GWh, electricity costs > $67M), Microsoft (annual energy consumption > 600GWh, electricity costs > $36M); (2) serious environmental pollution, IT carbon emissions accounted for 2% of global carbon emissions; (3) unexpected power outages, like Amazon experienced again in October 2012 due to power failure Power outage accident. [0003] Academia and industry have been improving data center energy efficiency (Energy Efficiency) thr...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the above defects or improvement needs of the prior art, the present invention provides an online control system for multiple energy sources in data centers under dynamic cloud service requests, the purpose of which is to solve the long-term operating expenses of the data center energy supply system existing in the existing system Large, unstable supply of new energy, fluctuations in electricity prices, need to obtain system data in advance or problems with steady-state distribution

Images