Cloud Computing & HPC – Synergy or Competition?

Chairs:

Prof. Dr. Dieter Kranzlmüller, Full Professor at the Ludwig-Maximilians-Universität München (LMU) & Member of the Board of the Leibniz Supercomputing Centre (LRZ), Germany

Dr. Martin Antony Walker, Independent Consultant, France

Cloud computing has the potential to become a major tool for scientists and engineers to access computing and storage. Such a situation is the logical outcome of ongoing developments in the technology and economics of information and communication technologies. These developments are driving the evolution of new ways of resourcing computing and storage across many, if not all, application domains.

Cloud computing is the current stage of a long-term evolution of information technology towards computing as a utility service. Computing clouds are enabled through automation and virtualization technologies that make it possible to abstract away much of the complexity of accessing vast amounts of computing and storage. Cloud providers can pass on the economies of scale (and green virtues) of huge centrally managed and operated data centers through a straightforward "pay for what you use" business model. As such, cloud computing is currently the hottest topic in scalable distributed computing, potentially enabling anyone with a credit card to access as much computing and storage as they need without having to acquire, support, or maintain the underlying hardware.

Current cloud computing vendors offer access to loosely connected collections of industry-standard virtual machines, and pools of storage. A large, and growing, number of companies propose tools and environments to simplify the ability to exploit computing and storage "in the cloud". There are community discussions about interoperability in the cloud, although so far no standards, either between different service levels within a single cloud, or among different clouds, have emerged.

The major concern in high-performance computing is performance. Primary determinants of performance are data locality and locality of reference, and communications latency and bandwidth. While there is no reason in principle why cloud computing vendors could not make these determinants selectable by cloud users, none of them is accessible through cloud computing offerings today.

The purpose of this session is to assess current implementations and deployments of cloud technology, what users are doing with cloud computing today, and most importantly for the audience of ISC'09, to assess the possibility of high-performance computing in the cloud. The session will provide key insights for HPC practitioners, observers, analysts, and vendors interested in scalable and distributed computing.

This session consists of two parts, separated by a one-hour break. The session will be opened with the chairmen setting the stage with a brief exposition of the context. The first part will contain presentations from three major IT vendors, HP, IBM and Sun Microsystems, followed by perspectives from a supercomputing center, JSC. The second part will start with the presentation from a major software house, Microsoft, and continue with the presentations from three IT vendors driving cloud computing today, Google, Amazon, and Yahoo!, followed by a round table panel discussion to field questions from the audience.

The eight speakers are outstanding leaders in the fields of high-performance computing and cloud computing. The session will conclude with a round table to encourage the active participation of attendees.

The speakers are:

• Richard Kaufmann
  CTO Scalable Computing Infrastructure Organization, Hewlett-Packard, USA
• Marc Hamilton
  VP HPC & Cloud Computing, Sun Microsystems, USA
• Prof. Dr. Dr. Thomas Lippert
  Director, Jülich Supercomputing Center (JSC)
• Prof. Dr. Dan Reed
  Scalable & Multicore Computing Strategist, Microsoft Research
• Robin Willamson
  Engineering Director, Google
• Simone Brunozzi
  Amazon Web Services Technology Evangelist – Europe, Amazon
• Dr. Sanjay Radia
  Senior Architect, Hadoop Project, Yahoo!, USA

The session is targeted at multiple groups each looking at cloud computing and HPC from a different angle: scientists who want up-to-date knowledge about the feasibility of HPC in cloud environments, system architects who want side-by-side comparisons of offerings from key vendors and providers, and finally IT professionals with budget authority who need to make decisions about future HPC and cloud computing environments in academia, research institutions, and industry. All will get a deeper understanding of the inherent complexity and state-of-the-art hardware and software components. The session will enable them to make technical and economic decisions based on facts and evidence from real life implementations.