Abstract
In the following paper we present the analysis of highly interconnected topologies like hypercube and torus and how they
can be implemented in data centers in order to cope with the rapid increase and demands for performance of the internal
traffic. By replicating the topologies in NS3 and subjecting them to uniformly distributed traffic routed by shortest path
algorithms we are able to extract statistics related to average throughput, latency and loss rate that show the decrease in the
average throughput per connection is only about 5% for the hypercube compared to 16% for the 3D torus when we scale the
network from 16 to 512 switches. The performance measurements are supported by abstract metrics that that also give a
cost and complexity indication in choosing the right topology for the required application.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of ICTON 2015 |
| Number of pages | 4 |
| Publication date | 2015 |
| Publication status | Published - 2015 |
| Event | 17th International Conference on Transparent Optical Networks - Danubius Thermal Hotel Helia, Budapest, Hungary Duration: 5 Jul 2015 → 9 Jul 2015 Conference number: 17 |
Conference
| Conference | 17th International Conference on Transparent Optical Networks |
|---|---|
| Number | 17 |
| Location | Danubius Thermal Hotel Helia |
| Country/Territory | Hungary |
| City | Budapest |
| Period | 05/07/2015 → 09/07/2015 |
Keywords
- NS3
- Hypercube
- Torus
- Interconnection networks
- Data Center topologies
- Performance evaluation