The performance of the Cedar multistage switching network
- Univ. of Illinois, Urbana, IL (United States). Center for Supercomputing Research and Developing
While multistage switching networks for vector multiprocessors have been studied extensively, detailed evaluations of their performance are rare. Indeed, analytical models, simulations with pseudo-synthetic loads, studies focused on average-value parameters, and measurements of networks disconnected from the machine, all provide limited information. In this paper, instead, the authors present an in-depth empirical analysis of a multistage switching network in a realistic setting: they use hardware probes to examine the performance of the omega network of the Cedar shared memory machine executing real applications. The machine is configured with 16 vector processors. The analysis suggests that the performance of multistage switching networks is limited by traffic non-uniformities. They identify two major non-uniformities that degrade Cedar`s performance and are likely to slow down other networks too. The first one is the contention caused by the return messages in a vector access as they converge from the memories to one processor port. This traffic convergence penalizes vector reads and, more importantly, causes tree saturation. The second non-uniformity is the uneven contention delays induced by even a relatively fair scheme to resolve message collisions. Based on their observations, their argue that intuitive optimizations for multistage switching networks may not be cost-effective. Instead, they suggest changes to increase the network bandwidth at the root of the traffic convergence tree and to delay traffic convergence up until the final stages of the network.
- OSTI ID:
- 87630
- Report Number(s):
- CONF-941118-; ISBN 0-8186-6605-6; TRN: IM9535%%262
- Resource Relation:
- Conference: Supercomputing `94 meeting, Washington, DC (United States), 14-18 Nov 1994; Other Information: PBD: 1994; Related Information: Is Part Of Supercomputing `94: Proceedings; PB: 849 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
SNAIL: A multiprocessor based on the simple serial synchronized multistage interconnection network architecture
Center for Supercomputing Research and Development (CEDAR). Progress report, January 25-December 24, 1986