The network architecture of the Connection Machine CM-5 (extended abstract)
SPAA '92 Proceedings of the fourth annual ACM symposium on Parallel algorithms and architectures
Universal schemes for parallel communication
STOC '81 Proceedings of the thirteenth annual ACM symposium on Theory of computing
Oblivious routing for fat-tree based system area networks with uncertain traffic demands
Proceedings of the 2007 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
VL2: a scalable and flexible data center network
Proceedings of the ACM SIGCOMM 2009 conference on Data communication
Hedera: dynamic flow scheduling for data center networks
NSDI'10 Proceedings of the 7th USENIX conference on Networked systems design and implementation
On the Data Path Performance of Leaf-Spine Datacenter Fabrics
HOTI '13 Proceedings of the 2013 IEEE 21st Annual Symposium on High-Performance Interconnects
Hi-index | 0.00 |
Performance optimized datacenters (PoDs) require efficient PoD interconnects to deal with the increasing volumes of inter-server (east-west) traffic. To cope with these stringent traffic patterns, datacenter networks are abandoning the oversubscribed topologies of the past, and move towards full-bisection fat-tree fabrics. However, these fabrics typically employ either single-path or coarse-grained (flow-level) multi-path routing. In this paper, we use computer simulations and analysis to characterize the waste of bandwidth that is due to routing inefficiencies. Our analysis suggests that, under a randomly selected permutation, the expected throughputs of d-mod-k routing and of flow-level multi-path routing are close to 63% and 47%, respectively. Furthermore, nearly 30% of the flows are expected to undergo an unnecessary 3-fold slowdown. By contrast, packet-level multi-path routing consistently delivers full throughput to all flows, and proactively avoids internal hotspots, thus serving better the growing demands of inter-server (east-west) traffic.