Timo Schneider, J. Dinan, M. Flajslik, K. D. Underwood, and Torsten Hoefler:
Fast Networks and Slow Memories: A Mechanism for Mitigating Bandwidth Mismatches
(In Proceedings of the 25th Annual Symposium on High-Performance Interconnects (HOTI'17), Aug. 2017)
Abstract
The advent of non-volatile memory (NVM) technologies has added an interesting nuance to the node level memory
hierarchy. With modern 100 Gb/s networks, the NVM tier of storage can often be slower than the high performance network in the
system; thus, a new challenge arises in the datacenter. Whereas prior efforts have studied the impacts of multiple sources targeting
one node (i.e., incast) and have studied multiple flows causing congestion in inter-switch links, it is now possible for a single flow from a
single source to overwhelm the bandwidth of a key portion of the memory hierarchy. This can subsequently spread to the switches and
lead to congestion trees in a flow-controlled network or excessive packet drops without flow control. In this work we describe protocols
which avoid overwhelming the receiver in the case of a source/sink rate mismatch. We design our protocols on top of Portals 4, which
enables us to make use of network offload. Our protocol yields up to 4x higher throughput in a 5k node Dragonfly topology for a
permutation traffic pattern in which only 1% of all nodes have a memory write-bandwidth limitation of 1/8th of the network bandwidth.
Documents
download article:
Recorded talk (best effort)
BibTeX
@inproceedings{, author={Timo Schneider and J. Dinan and M. Flajslik and K. D. Underwood and and Torsten Hoefler}, title={{Fast Networks and Slow Memories: A Mechanism for Mitigating Bandwidth Mismatches}}, year={2017}, month={Aug.}, booktitle={Proceedings of the 25th Annual Symposium on High-Performance Interconnects (HOTI'17)}, source={http://www.unixer.de/~htor/publications/}, }
