SciNet: Lessons Learned from Building a Power-efficient Top-20 System and Data Centre

By C. Loken, D. Gruner, L. Groer, R. Peltier, Neil Lawren Bunn, M. Craig, Teresa Henriques, J. Dempsey, Ching-Hsing Yu, Joseph M. Chen and 6 more. 2010

501 citations recorded by the source corpus.

SciNet, one of seven regional HPC consortia operating under the Compute Canada umbrella, runs Canada's first and third fastest computers (as of June 2010) in a state-of-the-art, highly energy-efficient datacentre with a Power Usage Effectiveness (PUE) design-point of 1.16. Power efficiency, computational "bang for the buck" and system capability for a handful of flagship science projects were important criteria in choosing the nature of the computers and the data centre itself. Here we outline some of the lessons learned in putting together the systems and the data centre that hosts Canada's fastest computer to date.

Subfields: Computer Networks and Communications, Hardware and Architecture, Nuclear and High Energy Physics

View Paper: https://www.semanticscholar.org/paper/b3cb061312886a0832450236a3e6d1beaa00a7d1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

1. Joaquim R. Martins
2. Timothy P. Lillicrap

In Hardware and Architecture

1. Joaquim R. Martins
2. Gaetan K. W. Kenway

Citations in the corpus

• Expected drag minimization for aerodynamic design optimization based on aircraft operational data (2017, methodology): “All the cases are run in parallel [27]: the L2 cases are performed with 32 processors, while the L1 cases use 480 processors.”
• Towards deep learning with segregated dendrites (2016, methodology): “Some of the simulations were run on the SciNet High-Performance Computing platform (Loken et al., 2010).”