Title: No Power, No Cloud
Speaker: Christian Belady, P.E.
Director of Hardware Architecture - Partner, Extreme Computing Group, Microsoft Research

Abstract:

This presentation will provide an overview of the trends and challenges the industry is facing with respect to the growth in the cloud infrastructure. Among those challenges that will be discussed are the high infrastructure costs and huge power demands by large scale data centers. While these are formidable challenges, there are still huge opportunities to rethink how the industry will build the cloud infrastructure of the future. This presentation will discuss some of the opportunities you can leverage today and share examples on how Microsoft is tackling these problems today and tomorrow.

Short Bio:

A recognized leader in the industry, Christian Belady is the Director of Hardware Architecture in the Extreme Computing Group at Microsoft Research where he leads a team responsible for exploring hardware opportunities related to the future of client plus cloud computing. Prior to his current role, he was Microsoft’s Principal Infrastructure Architect for Global Foundation Services where his role was to improve both efficiency and cost in their online services infrastructure. His responsibilities included driving initiatives for sustainability in the data center and infrastructure space, and he was one of the key architects for the Generation 4 modular data centers. Prior to joining Microsoft, Christian was a Distinguished Technologist for HP where his responsibilities included driving the technology direction in HP's server products and their environments, as well as driving industry data center initiatives. In addition, his earlier employers include Convex Computers (acquired by HP), TI and IBM. With over 78 US patents and many international patents, Christian is an ASME and IMAPS Fellow and a founding member of ASHRAE's TC9.9, which is responsible for developing data center guidelines. He was one of the early architects of the Green Grid and currently the treasurer for the organization, and is an originator of the Power Usage Effectiveness (PUE) metric used to determine the energy efficiency of a data center which is now broadly used by the industry. He also works closely with government agencies globally to define efficiency metrics for data centers and servers. Christian has published several dozen papers, is frequently quoted in the press, and is a featured speaker on power and cooling trends in the industry. Since the late 1990s, the focus of his publications has been on data centers and the industry’s need for engineering efficient computing environments. Christian holds engineering degrees from Cornell University (BS) and Rensselaer Polytechnic Institute (MS) and a business degree from the University of Texas at Dallas (MA) where he was honored with the 2010 Distinguished Alumni Award.

Title: Scaling Storage into the Exascale Era
Speaker: Garth Gibson
Professor, Carnegie Mellon University, and
Co-Founder & CTO, Panasas Inc.

Abstract:

The path of storage technology from terascale to petascale focused on effective parallel access to massive data. Reaching for exascale, compute node core counts are exploding and solid state disks are infiltrating storage while magnetic disks continue to grow capacity far faster than bandwidth and random access performance is essentially flat. In this talk I will share my excitement for the coming decade of storage system challenges and innovations.

Short Bio:

I joined the faculty of CMU's Computer Science Department in 1991. Previously I received a Ph.D. and a M.Sc. in Computer Science in 1991 and 1987, respectively, from the University of California at Berkeley. Prior to Berkeley, I received a Bachelor of Mathematics in Computer Science and Applied Mathematics in 1983 from the University of Waterloo in Ontario, Canada. In 1993 I founded CMU's Parallel Data Laboratory (PDL) and led it until April 1999. Today the PDL is led by Greg Ganger. The PDL is a community that typically comprises between 3 to 6 faculty, 1 to 2 dozen students and 4 to 10 staff. It receives support and guidance from a consortium of 10 to 20 companies with interests in storage systems, the Parallel Data Consortium. This community holds biannualretreats and workshops to exchange technology ideas, analysis and future directions. The publications of the PDL are available for your inspection. The principal contributions of my last twenty years of research Redundant Arrays of Inexpensive Disks (RAID), Informed Prefetching and Caching (TIP) and Network-Attached Secure Disks (NASD), now an ANSI SCSI command set standard (OSD), have all stimulated derivative research and development in academia and industry. RAID, in particular, is now the organizing concept of a 10+ billion-dollar marketplace (more on RAID in my 1995 RAID tutorial). In 1999 I started Panasas Inc., a scalable storage cluster company using an object storage architecture and providing 100s of TB of high-performance storage in a single management domain for national laboratory, energy sector, auto/aero-design, life sciences, financial modeling, digital animation, and engineering design markets. In 2006 I founded a Petascale Data Storage Institute (PDSI) for the Department of Energy's Scientific Discovery through Advanced Computing (SciDAC). Led by CMU, with partners at Los Alamos, Sandia, Oak Ridge, Pacific Northwest and Lawrence Berkeley National Labs, and University of California, Santa Cruz and University of Michigan, Ann Arbor, this Institute gathers together leading experts in leadership class supercomputing storage systems to address the challenges involved in moving from today's terascale computers to the petascale computers of the next decade. In 2008 I turned to Data Intensive Scalable Computing, Clouds, and Scalable Analytics, participating in the design and installation of 2 TF, 2TB, 1/2PB of computing in an OpenCirrus and an OpenCloud cluster.

Keynote 3, Thursday, September 23, 2010