Abstract
We observe and analyze usage of the login nodes of the leadership class Summit supercomputer from the perspective of an ordinary user—not a system administrator—by periodically sampling user activities (job queues, running processes, etc.) for two full years (2020–2021). Our findings unveil key usage patterns that evidence misuse of the system, including gaming the policies, impairing I/O performance, and using login nodes as a sole computing resource. Our analysis highlights observed patterns for the execution of complex computations (workflows), which are key for processing large-scale applications.
Original language | English |
---|---|
Title of host publication | Computational Science - ICCS 2022, 22nd International Conference, Proceedings |
Editors | Derek Groen, Clélia de Mulatier, Valeria V. Krzhizhanovskaya, Peter M.A. Sloot, Maciej Paszynski, Jack J. Dongarra |
Publisher | Springer Science and Business Media Deutschland GmbH |
Pages | 516-529 |
Number of pages | 14 |
ISBN (Print) | 9783031087509 |
DOIs | |
State | Published - 2022 |
Event | 22nd Annual International Conference on Computational Science, ICCS 2022 - London, United Kingdom Duration: Jun 21 2022 → Jun 23 2022 |
Publication series
Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|
Volume | 13350 LNCS |
ISSN (Print) | 0302-9743 |
ISSN (Electronic) | 1611-3349 |
Conference
Conference | 22nd Annual International Conference on Computational Science, ICCS 2022 |
---|---|
Country/Territory | United Kingdom |
City | London |
Period | 06/21/22 → 06/23/22 |
Funding
This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The publisher acknowledges the US government license to provide public access under the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). Acknowledgments. This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725. We acknowledge Suzanne Parete-Koon for early brainstorming of some of the ideas presented here. We thank Scott Atchley, Bronson Messer, and Sarp Oral for their thorough revision of this paper.
Keywords
- High Performance Computing
- User behavior
- Workload characterization