Routing over Parallel Queues with Time Varying Channels with Application to Satellite and Wireless Networks

We consider the problem of routing packets from an arbitrary input stream X(t) over a collection of heterogeneous queues in parallel. When the processing rates (μ1,..., μn) of the queues are constant, a simple work conserving routing strategy πWC is shown to hold total system backlog within a fixed upper bound from the resulting backlog of any other policy. Similar results apply to systems with time varying processing rates (μ1(t),..., μn(t)) when routing decisions can be postponed by placing packets in a pre-queue. For the case when routing decisions must be made immediately upon arrival, we demonstrate that the nonpredictive policy of routing packets to the shortest queue ensures system stability (and maintains low packet loss rates for finite buffer systems) whenever the system is stabilizable. Finally, we consider a joint problem of routing and power allocation where, for each time varying channel state ci(t), the rate of each queue i can be varied by adjusting a power parameter pi (subject to power constraints) according to a rate-power curve μi(ci, pi). A throughput maximizing algorithm is developed for this joint problem.

[1]  Anthony Ephremides,et al.  Extension of the optimality of the threshold policy in heterogeneous multiserver queueing systems , 1988 .

[2]  Morton J. M. Posner,et al.  A level-crossing approach to the solution of the shortest-queue problem , 1997, Oper. Res. Lett..

[3]  Bruce Hajek,et al.  Scheduling with asynchronous service opportunities with applications to multiple satellite systems , 1993, IEEE Trans. Autom. Control..

[4]  Frank Thomson Leighton,et al.  Methods for message routing in parallel machines , 1992, STOC '92.

[5]  David R. Karger,et al.  A better algorithm for an ancient scheduling problem , 1994, SODA '94.

[6]  Edward G. Coffman,et al.  An Application of Bin-Packing to Multiprocessor Scheduling , 1978, SIAM J. Comput..

[7]  Amos Fiat,et al.  New algorithms for an ancient scheduling problem , 1992, STOC '92.

[8]  Leandros Tassiulas,et al.  Dynamic server allocation to parallel queues with randomly varying connectivity , 1993, IEEE Trans. Inf. Theory.

[9]  Ronald L. Graham,et al.  Bounds on Multiprocessing Timing Anomalies , 1969, SIAM Journal of Applied Mathematics.

[10]  David P. Williamson,et al.  Scheduling parallel machines on-line , 1991, [1991] Proceedings 32nd Annual Symposium of Foundations of Computer Science.

[11]  Leandros Tassiulas,et al.  Stability properties of constrained queueing systems and scheduling policies for maximum throughput in multihop radio networks , 1990, 29th IEEE Conference on Decision and Control.

[12]  B. Hajek Optimal control of two interacting service stations , 1982, 1982 21st IEEE Conference on Decision and Control.

[13]  Eytan Modiano,et al.  Power and server allocation in a multi-beam satellite with time varying channels , 2002, Proceedings.Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies.

[14]  Eytan Modiano,et al.  Convexity and optimal load distributions in work conserving */*/1 queues , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[15]  Jean C. Walrand,et al.  An introduction to queueing networks , 1989, Prentice Hall International editions.

[16]  Anthony Ephremides,et al.  A simple dynamic routing problem , 1980 .