We describe a new framework for resource allocation that unifies the well-known proportional share and resource reservation policies. Each client is characterized by two parameters: a weight that represents the rate at which the client "pays" for the resource, and a share that represents the fraction of the resource that the client should receive. A fixed rate corresponds to a proportional share allocation, while a fixed share corresponds to a reservation. Furthermore, rates and shares are duals of each other. Once one parameter is fixed the other becomes fixed as well. If a client asks for a fixed share then the level of competition for the resource determines the rate at which it has to pay, while if the the rate is fixed, level of competition determines the service time the client should receive. To implement this framework we use a new proportional share algorithm, called Earliest Eligible Virtual Deadline First, that achieves optimal accuracy in the rates at which process execute...

We describe a new approach to implement proportional share resource allocation and to provide different levels of service quality. We consider multiple clients that compete for a time-shared resource, and we associate to each client a certain amount of funds. At the beginning of every timeslice a client is selected and it is granted to use the resource during that time-slice. For selecting a client, we use a new deterministic scheme that allocates time-slices to every client in proportion to its funds. More precisely, we prove that our scheme ensures that out of m consecutive time-slices that are allocated to all clients, the difference between the actual number of time-slices allocated to a client and the expected one is at most proportional to log m. The algorithm is extended in order to support different levels of service quality; at the highest level it guarantees that a client receives a share of the resource that is both superior and inferior bounded. 1 Introduction One of the m...