A Framework For Truthful Online Auctions In Cloud Computing With Heterogeneous User Demands

The paradigm of cloud computing has spontaneously prompted a wide interest inmarket-based resource allocation mechanisms by which a cloud provider aims at efficientlyallocating cloud resources among potential users. Among these mechanisms, auction-stylepricing policies, as they can effectively reflect the underlying trends in demand and supplyfor the computing resources, have attracted a research interest recently. However, after anextensive survey of the current purchasing designs in cloud market, we found that most ofthem suffer from the following drawbacks:(1).unable to timely reflect the supply-demandfluctuations;(2).unable to fulfill heterogeneous user requirements (3).vulnerable to users’cheating behaviors. This paper conducts the first work on a framework for truthful onlinecloud auctions where users with heterogeneous demands could come and leave on thefly. Our framework desirably supports a variety of design requirements, including (1)dynamic design for timely reflecting fluctuation of supply-demand relations,(2) jointdesign for supporting the heterogeneous user demands, and (3) truthful design for dis-couraging bidders from cheating behaviors. Concretely speaking, we first design a novelbidding language, wherein users’ heterogeneous demands are generalized to regulate andconsistent forms. Besides, building on top of our bidding language we propose COCA, anincentive-Compatible (truthful) Online Cloud Auction mechanism based on two proposedguidelines. Our theoretical analysis shows that the worst-case performance of COCA canbe well-bounded. Plus, our further discussion shows that COCA performs quite well whensome other factors concerned in online auction design are taken into consideration. Finally,in simulations the performance of COCA is seen to be comparable to the well-knownoff-line Vickrey-Clarke-Groves (VCG) mechanism[1].