Project Overview

The goal of this three-year project is to deliver an operating system for future extreme-scale parallel computing platforms that addresses the major technical challenges of energy efficiency, managing massive parallelism and deep memory hierarchies, and providing resilience in the presence of increasing failures.

Hobbes is a collaboration between 4 national labs and 8 universities. Full project details are available at xstack.sandia.gov/hobbes. The UA contribution focuses on developing models for power, as well as models that integrate power and resilience.

Supported by Sandia National Laboratory and DOE under subcontract number 1399016. Any opinions, findings, and conclusions expressed are those of the author(s) and do not necessarily reflect the views of Sandia National Laboratory or DOE.

People
Faculty
David K. Lowenthal
Graduate Students
Peter Bailey
Publications
Adaptive Configuration Selection for Power-Constrained Heterogeneous Systems
P. Bailey, D.K. Lowenthal, V. Ravi, B. de Supinski, B. Rountree, M. Schulz
ICPP, September 2014