Toward a More Realistic Performance Evaluation of Interconnection Networks

• Authors: Walter B. Ligon, III;Umakishore Ramachandran
• Venue: IEEE Transactions on Parallel and Distributed Systems
• Year: 1997

Interconnection network design plays a central role in the design of parallel systems. Most of the previous research has evaluated the performance of interconnection networks in isolation. In this study, we investigate the relationship between application program characteristics and interconnection network performance using an execution driven simulation testbed: the Reconfigurable Architecture Workbench (RAW). We simulate five topological configurations of a k-ary n-cube interconnect and four different network link models for a 4,096 node SIMD machine, and quantify the impact of the network on two application programs. We provide experimental evidence that such "in-context" simulation provides a better view of the impact of network design variables on system performance. We show that recent results, indicating that low-dimensional designs provide better ICN performance, ignore application requirements that may favor high-dimensional designs. Furthermore, applications that would appear to favor low-dimensional designs may not, in fact, be significantly impacted by the network's dimensionality. We experimentally test the results of published performance models comparing the use of a synthetic load to that of a load generated by a typical application program. The experiments indicate that the standard metric of average message latency can vary considerably under different application loads and that average message latency may not reflect overall application performance.In particular, at the level of the offered (application generated) load to the network, the topological properties of the network are important in determining the average message latency. However, for overall application performance, we found that the network topology may not be critical so long as there is sufficient network bandwidth. In such cases, the results suggest that optimizing the implementation cost of the network should be the key design criterion. We also present a simple abstraction for the network that captures all the important design parameters of the interconnect that can be easily incorporated into any execution-driven simulation framework.