Daily Coverage From
LinuxWorld Conference & Expo
San Francisco - 2006
Brought to You by HPCwire and Tabor Communications
Go to HPCwire Go to Myricom Go to GRIDtoday


August 17, 2006 - Features

The Fabric-Based Data Center

Sharad Mehrotra

In this Q&A with HPCwire, Sharad Mehrotra, President and CEO of Fabric7 Systems discusses how his company is addressing I/O virtualization as well as other data center computing challenges. Since launching their first products in November 2005, Fabric7 Systems has attracted a good deal of attention with its innovative commodity-based server solutions. Using AMD Opteron processors and the Linux or Windows operating system as the foundation of their "fabric" computing model, the company offers systems that provide features normally seen only in high-end enterprise servers: hardware partitioning and I/O virtualization. [more]


Linux Clusters Target Oil & Gas Applications

by S. Julio Friedmann
Carbon Management Program, Lawrence Livermore National Laboratory

The oil & gas industry operates in an increasingly challenging environment. The new challenges include more than high risk and high capital commitments, or declining fields and complex operations. One also wants to be smart. The good news is that smart is a lot cheaper than it used to be. Specifically, high performance computers are a lot less expensive than they used to be, and a lot more powerful. This article discusses how cutting-edge Linux clusters are tackling tough exploration, drilling and production problems for oil and gas companies. [more]


August 16, 2006 - Features

What HPC Networking Requires from the Linux Kernel

by Brice Goglin, Ph.D.
Myricom Inc.

Brice Goglin, Ph.D.

A number of people outside the core Linux kernel community have been modifying Linux in non-standard ways to provide more optimized high performance computing (HPC) networking solutions with specialty networks and network stacks. Brice Goglin of Myricom Inc. argues that there would be significant benefits to include a better framework for the specialized HPC networks in the base Linux code. These changes to Linux would make the corresponding network stacks more efficient and much easier to maintain. Goglin says that now might be a good time for the community to take on this task, inasmuch as a convergence between HPC networking and traditional networking is starting to take place. [more]

HPC Linux Cluster Showcase

Open Interfaces Enables Parallel Applications Debugging

Chris Gottbrath

The Message Passing Interface (MPI) is used on most distributed memory architectures, such as clusters, to support parallel computing. Debuggers, such as Etnus' TotalView, also rely on the interface to examine and control parallel applications during execution. On Thursday, August 17, at the LinuxWorld Conference, Chris Gottbrath, Product Manager for the Etnus TotalView debugger, will talk about Etnus' efforts in helping to develop open interfaces for MPI. His presentation will highlight Etnus' collaboration with the MPICH and Open MPI developers to extend these interfaces for the next generation of MPI. In this Q&A with HPCwire, Chris previews some of the topics he will be discussing. [more]


Wyeth Meets Bioinformatics Needs with HPC Linux Cluster

Recent years have seen extraordinary growth in the field of bioinformatics: Completion of several high-profile initiatives, including the Human Genome Project, have raised awareness and fueled interest in the discipline. With this upsurge has come an associated explosion in the volumes of data and need for powerful computational resources. Wyeth Bioinformatics needed an HPC solution that would allow them to analyze massive amounts of data, and cluster technology was well-suited for their needs. This article describes how Wyeth Bioinformatics came to select an HP Linux Cluster and the benefits that resulted. [more]

August 15, 2006 - Features

Making Virtualization Real: Revisiting Beowulf-Class Linux Cluster Architectures

by Bob Monkman
Director, Product Management for Scyld Software, Penguin Computing

Bob Monkman

As cluster use in enterprises grows, the need for high performance computing that scales on-demand to adapt to ever-changing workload requirements and provide optimal system utilization is also growing. These needs in turn have driven many useful innovations. However, there has remained a fundamental assumption that a cluster or grid configuration is provisioned as a static, disk-based, full operating system installation on every single server. In this article, Bob Monkman, Director of Product Management for Scyld Software, Penguin Computing, discusses an alternative approach -- Cluster Virtualization. [more]

IT as Fuel for the Innovation Engine

by Ian Foster
Director, Computation Institute at ANL/University of Chicago, Univa Corp.

Dr. Ian Foster

Ian Foster discusses his belief that we will see significant progress in the creation and application of IT infrastructures architected specifically to facilitate innovation, and a shift from thinking of IT solely as a cost center to recognizing IT as a value enabler. Foster argues that the most successful companies are increasingly focusing their attention on the process of innovation and on the empowerment of the knowledge worker -- or innovator -- in order to establish a competitive edge. Competitive pressures are particularly acute in industries where product design and development require the use of sophisticated IT for complex and computationally intensive simulation, design optimization and data analysis. [more]

Building Blocks & Standards for HPC and Linux Clusters

by Gary Tyreman
Vice President of the Open Cluster Group, Platform Computing

Gary Tyreman

It is no secret in the HPC community that Linux clusters are becoming increasingly popular with market share growing significantly over the last two years. This growth is only sustainable if Linux clusters continue to move into new areas of growth and unlock new market segments. The growing appeal of using Linux clusters in organizations often boils down to choice and the freedom to select and build a custom solution that meets individual needs. However, as Linux grows in popularity the element of choice is inhibiting broader market adoption because of the inherent lack of stability and consistency. In this article, Gary Tyreman, Vice President of the Open Cluster Group, Platform Computing, discusses these issues. [more]


Top of Page