The Business of Energy in U.S. Academic Research Computing
By Dr. Gregory B. Newby. April, 2014
Table of Contents with Links Below the Fold. See Spreadsheet Especially.
- Executive Summary
- Project Plan: Impetus; Opportunity/Problem Statement; Project Scope; High Level Timeline and Milestones; Metrics and Measurement of Outcomes; Resource Requirements
- Industry analysis: Is Academic Research Computing an Industry?; Academic Research Computing as an Industry; The High Performance Computing Industry; Return on Investment; CASC Perspective on the Industry; “Small HPC” BoF Perspective on Industry; Industry Maturity
- Systems within systems: The research computing environment: Systems Overview; Systems Characteristics for Research Computing; Systems in Research Computing Environments; Systems Components for Research Computing
- Key Resources
- Stakeholder analysis: Organizational Stakeholders; State & Regional Stakeholders; National and International Stakeholders
- Key Performance Indicators and Dashboards: From Mission and Vision to Goals
- Return on Investment
- The role of energy in research computing: Fundamentals of electrical usage for computing; The physics of electricity in computing; Cooling; Backup power; Energy and Data Movement; Cheap and Reliable Electricity; Large-scale Data Movement is to be Avoided; Local Data Movement is also to be Avoided
- Equipment lifecycles and technology refresh: Materials; Assessing Impact; Simple Measures; Moving Forward; Energy Roles in Procurement and Operation
- Organizational Culture: Introduction to Organizational Culture; CASC as a Culture Keeper for Research Computing; Stewardship as a Key Goal of Organizational Culture in Research Computing; Challenges of Maintaining Stewardship; Staffing for Research Computing; Employee Development
- Conclusion
- The Spreadsheet … also a CASC presentation from April 2014
- References and Notes
- Acknowledgements
Please email the author with any questions or suggestions.
Greg Newby is the Chief Technology Officer of Compute Canada, the national supercomputing organization. Newby is bringing advanced research computing in Canada towards a future of meeting all social and economic needs (including processing of an order of magnitude more information in near real time) while also migrating it toward open source foundations including open source security. Part of the goal is to render the system uncorruptible by human elements and shortfalls. In his previous position he managed the acquisition of the largest newest supercomputer for the King Abdullah University of Science & Technology in Saudi Arabia. Previously he was Director of the Arctic Region Supercomputing Center.at the University of Alaska Fairbanks. Full bio.