Express Cubes: Improving the Performance of k-ary n-cube Interconnection Networks
IEEE Transactions on Computers
Performance evaluation of a new routing strategy for irregular networks with source routing
Proceedings of the 14th international conference on Supercomputing
The Impact of Pipelined Channels on k-ary n-Cube Networks
IEEE Transactions on Parallel and Distributed Systems
ServerNet Deadlock Avoidance and Fractahedral Topologies
IPPS '96 Proceedings of the 10th International Parallel Processing Symposium
Improving Routing Performance in Myrinet Networks
IPDPS '00 Proceedings of the 14th International Symposium on Parallel and Distributed Processing
Applying In-Transit Buffers to Boost the Performance of Networks with Source Routing
IEEE Transactions on Computers
Distance constrained mapping to support NoC platforms based on source routing
Euro-Par'09 Proceedings of the 2009 international conference on Parallel processing
Hi-index | 0.01 |
Networks of workstations (NOWs) are becoming increasingly popular as a cost-effective alternative to parallel computers. In these machines, the network connects processors using irregular topologies, providing the wiring flexibility, scalability, and incremental expansion capability required in this environment. In addition, when performance is the primary concern, these network products are being used to build large commodity clusters with regular topologies [10]. In previous papers [3, 4], we have proposed the in-transit buffer mechanism to improve network performance, applying it to NOWs with irregular topology and source routing. This mechanism allows the use of minimal paths among all hosts, breaking cyclic dependencies between channels by storing and later re-injecting packets at some intermediate hosts. In this paper, we apply the in-transit buffer mechanism to regular networks with source routing in order to improve their performance. In addition, two path selection policies are evaluated. The first one will always choose the same minimal path from source to destination, whereas the second one will choose from different alternative minimal paths in a round-robin fashion. The evaluation results show that the overall network throughput can be doubled for large networks.