Skip to Content

Monet — Blue Waters Network Data

Data Set

Data set available for download.

Description

This dataset contains collected and aggregated network information from NCSA's Blue Waters system, which is comprised of 27,648 nodes connected via Cray Gemini* 3D torus (dimension 24x24x24) interconnect, from Jan/01/2017 to May/31/2017. Network performance counters for links are exposed via Cray's gpcdr (https://github.com/ovis-hpc/ovis/wiki/gpcdr-kernel-module) kernel module. Lightweight Distributed Metric Service ([LDMS](https://github.com/ovis-hpc/ovis)) is used to sample the performance counters at 60 second intervals. Please read "README.md" file.

Acknowledgment:

This dataset is collected as a part of the Blue Waters sustained-petascale computing project, which is supported by the National Science Foundation and the state of Illinois. Blue Waters is a joint effort of the University of Illinois at Urbana-Champaign and its National Center for Supercomputing Applications. This material is partially based upon work supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, under Award Number 2015- 02674.

Citation [data]:

Saurabh, Jha; Archit, Patke; Mike, Showerman; Jeremy, Enos; Greg, Bauer; Zbigniew, Kalbarczyk; Ravishankar, Iyer; William , Kramer (2019): Monet - Blue Waters Network Dataset. University of Illinois at Urbana-Champaign. https://doi.org/10.13012/B2IDB-2921318_V1

Citation [Paper]:

Saurabh Jha, Archit Patke, Benjamin Lim, Jim Brandt, Ann Gentile, Greg Bauer, Mike Showerman, Larry Kaplan, Zbigniew Kalbarczyk, William T Kramer, Ravishankar K. Iyer (2020). Measuring Congestion in High-Performance Datacenter Interconnects. 17th USENIX Symposium on Networked Systems Design and Implementation (NSDI).