Research On Datacenter Workload Management Strategy Towards Demand Response

With the rapid development of cloud computing technology,the energy consumption and carbon emissions caused by datacenters for computing and storage have become issues that cannot be ignored.In recent years,some IT companies with large datacenters are using renewable energy as part of their datacenter energy supply.At the same time,with the development of the smart grid,it provides very flexible interaction functions with its users.In order to improve reliability and reduce peak load pressure,the smart grid attempts to propose various types of demand response programs to encourage users to participate.As a high power demand user of power grid,the datacenter is a critical demand response participant.Datacenter participation in demand response programs of smart grids can help the grid to adjust the load while reducing its own power costs so as to save electricity costs.In this paper,the research of aiming at the high energy consumption of the datacenter and how the datacenter participates in demand response for reducing its power costs is mainly done as follows:In order to reduce the energy consumption of the datacenter and make the datacenter make full use of renewable energy,the paper combines the flexibility characteristics of the datacenter workload,and at the same time considers the computing energy consumption and cooling energy consumption of datacenter.We propose a renewable energy-aware datacenter workload management strategy.Through simulation experiments,we prove that the strategy proposed in this paper can maximize the use of renewable energy and minimize the consumption of brown energy.Focusing at how the datacenter participates in smart grid demand response in order to minimize the power cost of the datacenter and reduce the peak load pressure of the smart grid,we propose a two-stage datacenter demand response strategy.The first stage determines whether to participate in demand response by predicting real-time electricity prices of the power grid and incentive information will be sent to encourage users to participate in the program to help shave the peak load.In the second stage,the datacenter interacts with its users by allowing users to submit bid information by auction.Then the datacenter selects the tasks of the winning users to postpone processing with awards,which reduces the electricity costs of the datacenter and effectively meets the demand response requirements of the smart grid.In this way,the goals of both parties could be achieved.