The chaos router: a practical application of randomization in network routing
SPAA '90 Proceedings of the second annual ACM symposium on Parallel algorithms and architectures
An efficient, fully adaptive deadlock recovery scheme: DISHA
ISCA '95 Proceedings of the 22nd annual international symposium on Computer architecture
IEEE Transactions on Parallel and Distributed Systems
On deadlocks in interconnection networks
Proceedings of the 24th annual international symposium on Computer architecture
The Case for Chaotic Adaptive Routing
IEEE Transactions on Computers
Interconnection Networks: An Engineering Approach
Interconnection Networks: An Engineering Approach
Journal of Parallel and Distributed Computing
The Alpha 21364 Network Architecture
IEEE Micro
A Flow Control Mechanism to Avoid Message Deadlock in k-ary n-cube Networks
HIPC '97 Proceedings of the Fourth International Conference on High-Performance Computing
Principles and Practices of Interconnection Networks
Principles and Practices of Interconnection Networks
Evaluation of interconnection network performance under heavy non-uniform loads
ICA3PP'05 Proceedings of the 6th international conference on Algorithms and Architectures for Parallel Processing
Age-based packet arbitration in large-radix k-ary n-cubes
Proceedings of the 2007 ACM/IEEE conference on Supercomputing
Improving a fault-tolerant routing algorithm using detailed traffic analysis
HPCC'07 Proceedings of the Third international conference on High Performance Computing and Communications
| Hi-index | 0.00 |
The performance of an interconnection network is measured by two metrics: average latency and peak network throughput. Network throughput is the total number of packets delivered per unit of time.Most synthetic network loads consist of sources injecting at the same given rate, using traffic patterns such as random, permutations or hot spot, which reflect the distribution of packet destinations in many parallel applications. The network is assumed to be fair: all source nodes are able to inject at the same rate. This work will show such assumption is unfounded for most router proposals. All router designs exhibited significant network unfairness under non-uniform loads. Some routers are also unfair under random traffic patterns. At loads above saturation, if the channel utilization is uneven, the injection matrix will become uneven: packet at low used areas will be accepted at a higher rate that those at the busy areas.As synthetic traffic does not reflect the coupled nature of the traffic generated by parallel applications, the impact of this unfairness on application performance could not be measured. New synthetic loads need to be developed to better evaluate network response beyond saturation.