The documents distributed by this server have been provided by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a noncommercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.
Publications of SPCL
|T. Schneider, J. Dinan, M. Flajslik, K. D. Underwood, a. 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)
AbstractThe 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.
Recorded talk (best effort)