Abstract: Increasing utilization is goal we all strive for. However, too high utilization can compromise availability, as we may not have resources to restart tasks in case of hardware failure. In this talk, I present a technique called backup tasks, which we use in Google to ensure that we have enough resources to withstand hardware failures.
Abstract: This presentation will show how large shared memory systems containing multiple databases can be used for applications such as fraud detection and prevention that are not possible using multiple disk resident databases. The SGI Altix UV system architecture capable of thousand of sockets and petabytes of memory will be presented. The architecture of the SGI MOJO system capable of over a petaflop in two racks will also be presented.
Abstract: In this presentation will be presented our vision for Exascale architecture and the Performance / Watt ratio expected. With the growing demand for energy and use of Green IT concepts we will show how to optimize space and power consumption with the Bullx ultra-dense blades and Mobull containers.
Abstract: The scale and complexity of current and future supercomputers brings a new set of challenges for scientific application developers. In order to achieve high performance on petascale systems, application developers need a well balanced system with a programming environment and system software that can address and hide the issues of scale and complexity of high end HPC systems. In this talk I will discuss some of the Cray XE design choices to overcome these challenges, providing adaptive and productive supercomputing with scalable performance and production efficiency.
Bio: Dr. Luiz DeRose is a Senior Principal Engineer and the Programming Environments Director at Cray Inc, where he is responsible for the programming environment strategy for all Cray systems. Dr. DeRose has a Ph.D. in Computer Science from the University of Illinois at Urbana-Champaign. With more than 20 years of high performance computing experience and a deep knowledge of its programming environments, he has published more than 40 peer-review articles in scientific journals, conferences, and book chapters, primarily on the topics of compilers and tools for high performance computing.