A survey on discriminatory processor sharing

Abstract

The Discriminatory Processor Sharing (DPS) model is a multi-class generalization of the egalitarian Processor Sharing model. In the DPS model all jobs present in the system are served simultaneously at rates controlled by a vector of weights {g k > 0; k = 1, . . . , K }. If there are N k jobs of class k present in the system, k = 1, . . . , K , each class-k job is served at rate g k / K j=1 g j N j . The present article provides an overview of the analytical results for the DPS model. In particular, we focus on response times and numbers of jobs in the system.

Monotonicity Properties for Multi-Class Queueing Systems

“…The policies we are particularly interested in are GPS [10,32] and DPS [20,12,1], two popular nonanticipating policies in multi-class single-server queues. Let GP S(φ) (DP S(φ)) denote a GPS (DPS) discipline that assigns weight φ j to class j, with…”

“…For a normally-loaded system we perform numerical experiments that provide further insight into the performance of the α-fair policies. Finally, we consider a multi-class single-server queue for which we are especially interested in weighted time-sharing policies such as Discriminatory Processor Sharing (DPS) [20,12,1] and Generalized Processor Sharing (GPS) [10,32]. For a single server with two classes we obtain that the weighted mean number of users in a class is monotone for DPS and GPS with respect to the ratio of the weights.…”

Beyond processor sharing

“…For an extensive overview of the literature on DPS we refer to the recent survey [7] and to [21] for results on tail distributions.…”

Simulating Flow Level Bandwidth Sharing with Pareto distributed File Sizes

“…The processor sharing queue has been perhaps the most common model for bandwidth sharing of flows (having the same RTT) of data transfer over the Internet [4], along with the Discriminatory Processor Sharing queue [7] adapted to the case of unequal round trip times. Several research groups have examined its validity through simulations [17,14,11,13].…”