Bandwidth Steering in HPC Using Silicon Nanophotonics

As bytes-per-FLOP ratios continue to decline, communication is becoming a bottleneck for performance scaling. This paper describes bandwidth steering in HPC using emerging reconfigurable silicon photonic switches. We demonstrate that placing photonics in the lower layers of a hierarchical topology e...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:SC19: International Conference for High Performance Computing, Networking, Storage and Analysis s. 1 - 25
Hlavní autori: Michelogiannakis, George, Shen, Yiwen, Teh, Min Yee, Meng, Xiang, Aivazi, Benjamin, Groves, Taylor, Shalf, John, Glick, Madeleine, Ghobadi, Manya, Dennison, Larry, Bergman, Keren
Médium: Konferenčný príspevok..
Jazyk:English
Vydavateľské údaje: ACM 17.11.2019
Predmet:
Bandwidth
Hardware
Nanophotonics
Network topology
Optical network units
Optical switches
Power demand
Silicon photonics
Throughput
Topology
ISSN:2167-4337
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:As bytes-per-FLOP ratios continue to decline, communication is becoming a bottleneck for performance scaling. This paper describes bandwidth steering in HPC using emerging reconfigurable silicon photonic switches. We demonstrate that placing photonics in the lower layers of a hierarchical topology efficiently changes the connectivity and consequently allows operators to recover from system fragmentation that is otherwise hard to mitigate using common task placement strategies. Bandwidth steering enables efficient utilization of the higher layers of the topology and reduces cost with no performance penalties. In our simulations with a few thousand network endpoints, bandwidth steering reduces static power consumption per unit throughput by 36% and dynamic power consumption by 14% compared to a reference fat tree topology. Such improvements magnify as we taper the bandwidth of the upper network layer. In our hardware testbed, bandwidth steering improves total application execution time by 69%, unaffected by bandwidth tapering.
ISSN:2167-4337
DOI:10.1145/3295500.3356145