Journal
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
Volume 35, Issue 19, Pages -Publisher
WILEY
DOI: 10.1002/cpe.7587
Keywords
average packet latency; fat-tree; hardware cost; hypercube; interconnection network; path diversity; routing algorithm
Ask authors/readers for more resources
This paper proposes two hybrid topologies, CAT and MiCAT, based on fat-tree and hypercube, to reduce hardware costs in parallel supercomputers. The results show that CAT and MiCAT can save up to 87% switches and 80% links compared to fat-trees while maintaining higher path diversity.
Fat-tree topologies are widely used in interconnect network designs for parallel supercomputers. In the classic fat-tree, compute nodes are connected to leaf stage switches by links. Given a large number of compute nodes, many switches and links are required, resulting in high hardware costs. To solve this problem, this paper proposes two hybrid topologies, k-Cube k-Ary n-Tree (CAT) and Mirrored k-Cube k-Ary n-Tree (MiCAT), based on fat-tree and hypercube. Instead of connecting k compute nodes directly to a leaf switch, we connect a k-cube to the switch, and each switch in the k-cube part connects k compute nodes. That is, this k-cube consists of 2(k) - 1 switches and k(2(k)-1) compute nodes. We give the shortest path routing algorithms and evaluate the path diversity, cost, performance, and average packet latency of CAT and MiCAT. The results show that CAT and MiCAT can save up to 87% switches and 80% links in a large-scale parallel system, k=n=8 for example, compared to fat-trees, and meanwhile, both CAT and MiCAT have higher path diversities than fat-trees.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available