Dominant Resource Fairness with Time Constraints in Cloud Computing

Fair resource is a key building block of any shared computing system. Recently fair division theory has emerged as a promising approach for the allocation of multiple computational resources among users. While in reality users are not all present in the system simultaneously. Dominant Resource Fairness(DRF) satisfies SI, PE AND EF, but violates SP. We have introduced Dominant Resource Fairness with Time Constraints (TDRF), a fair sharing model that generalizes max-min fairness to multiple resource types. TDRF has lots of good properties, it satisfies DSI, PE and SP. Users can exchange resources which are user don not need for resources which are user need by exchange mechanism. TDRF think about the time factor to improve the overall effectiveness of the system. If users stay the system more longer, the resources which user contributes will be use more longer by other users. We construct TDRF mechanisms that provably satisfy properties, and analyze the performance.