Isolating Cluster Users for Performance and Predictability

AsiaBSDCon 2009 Paper Session.
Abstract: At The Aerospace Corporation, we run a large FreeBSD based computing cluster to support engineering applications. These applications come in all shapes, sizes, and qualities of implementation. To support them and our diverse userbase we have been searching for ways to isolate jobs from one another in ways that are more effective than Unix time sharing and more fine grained than allocating whole nodes to jobs. In this paper we discuss the problem space and our efforts so far. These efforts include implementation of partial file systems virtualization and CPU isolation using CPU sets.
Author: Brooks Davis is a Engineering Specialist in the Technical Computing Services subdivision of The Aerospace Corporation. He has been a FreeBSD user since 1994, a FreeBSD committer since 2001, and a core team member since 2006. He earned a Bachelors Degree in Computer Science from Harvey Mudd College in 1998. His computing interests include high performance computing, networking, security, mobility, and, of course, finding ways to use FreeBSD in all these areas. When not computing, he enjoys reading, cooking, brewing, trying to introduce people to the pleasures of Slow Food, and pounding on red-hot iron in his garage blacksmith shop.

Category:

Science & Technology

Tags:

• BSD
• AsiaBSDCon
• AsiaBSDCon2009
• FreeBSD
• high-performance
• computing
• cluster