Best-effort scheduling of (m,k)-firm real-time streams in multihop networks

Abstract

In this paper, we address the problem of best-effort scheduling of (m,k)-firm real-time streams in multihop networks. The existing solutions for the problem ignore scalability considerations because the solutions maintain a separate queue for each stream. In this context, we propose a scheduling algorithm, EDBP, which is scalable (fixed scheduling cost) with little degradation in performance. The proposed EDBP algorithm achieves this by allowing multiplexing of streams onto a fixed number of queues and by using the notion of a look-ahead window. In the EDBP algorithm, at any point of time, the best packet for transmission is selected based on the state of the stream combined together with the laxity of the packet. Our simulation studies show that the performance of EDBP is very close to that of DBP-M (a known algorithm for the problem) with a significant reduction in scheduling cost. © 2000 Springer-Verlag Berlin Heidelberg.