Principles and Practices of Interconnection Networks
Principles and Practices of Interconnection Networks
Technology-Driven, Highly-Scalable Dragonfly Topology
ISCA '08 Proceedings of the 35th Annual International Symposium on Computer Architecture
The PERCS High-Performance Interconnect
HOTI '10 Proceedings of the 2010 18th IEEE Symposium on High Performance Interconnects
Avoiding hot-spots on two-level direct networks
Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis
Cray cascade: a scalable HPC system based on a Dragonfly network
SC '12 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis
On-the-Fly Adaptive Routing in High-Radix Hierarchical Networks
ICPP '12 Proceedings of the 2012 41st International Conference on Parallel Processing
Improving performance of all-to-all communication through loop scheduling in PGAS environments
Proceedings of the 27th international ACM conference on International conference on supercomputing
Randomizing task placement does not randomize traffic (enough)
Proceedings of the 2013 Interconnection Network Architecture: On-Chip, Multi-Chip
| Hi-index | 0.00 |
Dragonfly topologies are recent network designs that are considered one of the most promising interconnect options for Exascale systems. They offer a low diameter and low network cost, but do so at the expense of path diversity, which makes them vulnerable to certain adversarial traffic patterns. Indirect routing approaches can alleviate the performance degradation that these workloads experience. However, there are limits to the improvements that can be achieved using the indirect routing approach that is popular today, limits that are inherent to the Dragonfly topological structure. In this work, we explore these limits by providing a theoretical justification to why adversarial traffic patterns routed indirectly with an algorithm that perfectly distributes load across inter-Dragonfly-group links can still induce significant bottlenecks in the intra-group links. We equally provide estimations of the performance impact of these imbalances, as well as present a set of simulation based benchmarks that confirm the theoretical predictions for practical Dragonfly systems.