This will be a great meetup for those interested in high-performance storage for HPC! It's being hosted at MIT by meetup member Kurt Keville.
Torben Kling Petersen, from Xyratex Ltd, will be presenting on HPC File Systems.
Dr. Petersen has worked with high performance computing in one form or another since 1994 and is currently the Principal Solutions Architect for the Xyratex HPC Solutions Group. Over the years, both in Sun and subsequently in Oracle, he has worked in a number of capacities as lead architect for enterprise datacenter infrastructure for several telecommunication OEMs, technical lead for IPTV and product specialist for high-end visualization, and global architect for the Lustre BdM team as well as a Cloud Computing infrastructure architect to mention a few. In these roles, Torben has worked on many large HPC systems around the world including Sandia Red Sky in the US, ANU/BOM in Australia, CHPC in South Africa, ETH Zurich and many others in Europe. In addition to compute and storage systems, Torben has also worked on data center design including the new CSCS datacenter in Lugano, Switzerland. Within Xyratex, Torben has worked as the principal solutions architect for 2 years and has worked on every aspect of the ClusterStor product line. Dr. Peterson majored in Marine Biology and Biochemisty with his doctorate from Goteborg University, Sweden.
Sponsor: Computational Research in Boston and Beyond Seminar
I'll be attending and I'll be traveling quite a way to attend. I'm interested in hearing about any experiences people have with setting up their own scalable, Open Source, fault tolerant systems. I've been looking into SAS solutions with OpenSFS, ZFS (ZoL) and other things for our HPC environment. I'm not averse to non-Open Source products but it can be a strong impediment with our funding model.
If anyone with experience would care to talk about this during or after the meeting I'd very much like to make contact and talk. Or if people want to talk/email offline let me know.