Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network

  Advanced Search  

A Quantitative Comparison of Scheduling Algorithms for InputQueued Switches \Lambda

Summary: A Quantitative Comparison of Scheduling Algorithms
for Input­Queued Switches \Lambda
Nick McKeown Thomas E. Anderson
This paper quantitatively evaluates several alternative approaches to the scheduling of cells in a high­
bandwidth input­queued ATM switch. In particular, we compare the performance of three algorithms described
previously: FIFO queueing, parallel iterative matching (PIM), maximum matching and two new algorithms:
iterative round­robin matching with slip (SLIP) and least­recently used (LRU). For the synthetic workloads
we consider, including uniform and bursty traffic, SLIP performs almost identically to the other algorithms.
Cases for which PIM and SLIP perform poorly are presented, indicating that care should be taken when using
these algorithms. But, we show that the implementation complexity of SLIP is an order of magnitude less
than for PIM, making it feasible to implement a 32x32 switch scheduler for SLIP on a single chip.
1 Introduction
The past few years has seen increasing interest in arbitrary topology cell­based local area networks, such as
ATM [5]. In these networks, hosts are connected together by an arbitrary graph of communication links and
switches, instead of via a bus, as in Ethernet[20], or a ring, as in FDDI[4]. One reason for the popularity of
arbitrary topology networks is that they offer a number of potential advantages relative to other approaches[2]:
(i) aggregate throughput that can be much larger than that of a single link, (ii) the ability to add throughput
incrementally as the workload changes by simply adding extra links and switches, (iii) improved fault tolerance
by allowing redundant paths between hosts, (iv) and reduced latency because control over the entire network is


Source: Anderson, Tom - Department of Computer Science and Engineering, University of Washington at Seattle


Collections: Computer Technologies and Information Sciences