CloudHeat

Abstract

Datacenters are major energy consumers and dissipate an enormous amount of waste heat. Simple outdoor discharging of datacenter heat is energy-consuming and environmentally unfriendly. By harvesting datacenter waste heat and selling to the district heating system (DHS), both energy cost compensation and environment protection can be achieved. To realize such benefits in practice, an efficient market mechanism is required to incentivize the participation of datacenters. This work proposes CloudHeat, an online reverse auction mechanism for the DHS to solicit heat bids from datacenters. To minimize long-term social operational cost of the DHS and the datacenters, we apply a RFHC approach for decomposing the long-term problem into a series of one-round auctions, guaranteeing a small loss in competitive ratio. The one-round optimization is still NP-hard, and we employ a randomized auction framework to simultaneously guarantee truthfulness, polynomial running time, and an approximation ratio of 2. The performance of CloudHeat is validated through theoretical analysis and trace-driven simulation studies.