摘要
In this paper, we develop a unified approach for stochastic load balancing onvarious multiserver systems. We expand the four partial orderings defined inMarshall and Olkin, by defining a new ordering based on the set of functions thatare symmetric, L-subadditive and convex in each variable. This new partial orderingis shown to be equivalent to the previous four orderings for comparing deterministicvectors but differs for random vectors. Sample-path criteria and a probabilityenumeration method for the new stochastic ordering are established and the orderingis applied to various fork-join queues, routing and scheduling problems. Our resultsgeneralize previous work and can be extended to multivariate stochastic majorizationwhich includes tandem queues and queues with finite buffers.