INDUSTRY

By Steve Fisher, Editor In Chief -- In addition to a recent product win at Fermilab, the people of Linux NetworX have been busy. Most notably they have, here at SC2001, unveiled their ICE Box cluster hardware appliance. To learn more about the company’s recent endeavors Supercomputing Online interviewed Linux NetworX’s CTO Joshua Harr. Supercomputing: Joshua, thank you for the interview. How's your show been thus far? HARR: Supercomputing is always an excellent venue for us because the attendees are interested and knowledgeable about clusters. For example, a survey from last year’s show indicated that 78 percent of attendees where very interested in clusters—We expect that this year will be considerably higher. This is a great opportunity for Linux NetworX to demonstrate our expertise in clustering to the supercomputing industry Supercomputing: What are a couple major foci, what's new and exciting here at this year's show? How about for the year ahead? HARR: This is a very exciting show for Linux NetworX because we are launching ICE Box, our new cluster management hardware appliance. With Linux clusters becoming the high performance compute engine of choice, cluster management is playing an increasingly important role in the future adoption of this technology. Administrators need to know about the tools available to help them effectively manage their clusters. For the year ahead, Linux NetworX will continue to focus on cluster management so organizations can focus on their core competencies instead of spending valuable resources integrating, configuring and managing such systems. We will also be investing more resources to expand our professional services division, such as application porting and Linux cluster training and certification. Supercomputing: What do you feel is the primary "edge" ICE Box has over the competition? HARR: ICE Box is the only appliance of its kind designed specifically to improve the manageability of Linux clusters. By providing direct serial access to individual nodes within the cluster, ICE Box delivers a level of control, convenience and manageability not previously available for Linux cluster systems. Features like node health and environmental monitoring, power control, node reset capabilities, and advanced serial switching. These are all features that haven’t been previously available in one box, tailored specifically for Linux clustering. Supercomputing: In what market segments do you see ICE Box being the most successful...life sciences? Academics computing research? Government? Entertainment? HARR: The ICE Box was designed so it can be successful for any organization requiring Linux clusters, however, Linux NetworX specializes in providing clustering services to scientific research and defense, oil & gas, bioinformatics, visual effects/entertainment, and mechanical & electrical design. We’ve engineered our management tools, ICE Box and ClusterworX with flexibility to work in multiple environments and applications. Any organization that wants to dedicate their resources to high performance computing rather than system administration will benefit from ICE Box. Supercomputing: Any ICE Box early adopters you'd care to mention? Any recent customer successes? HARR: Although we’ve just officially launched the ICE Box product at SC 2001, we do have some customers that are using ICE Box. Lawrence Livermore National Lab and Tularik, a drug discovery and development company are using ICE Box with much success. For a customer testimonial about ICE Box from Tularik, please visit www.linuxnetworx.com/news/10.30.2001.49-Linux_NetworX_t.html For a customer testimonial about ICE Box from Lawrence Livermore, please visit www.linuxnetworx.com/news/8.28.2001.49-Lawrence_Liverm.html Supercomputing: Do you feel that cluster computing will eventually replace the stand-alone supercomputer? HARR: Linux clusters have become the high performance compute (HPC) engine of choice for many industries seeking raw number crunching power with greater flexibility, reliability, scalability and price/performance over traditional supercomputers. These factors are also becoming increasingly important for organizations with supercomputing needs. They don’t want to spend millions of dollars on a supercomputer that will only have to tossed aside in a couple of years for something more powerful. The scalability of clusters make them the obvious choice for organizations that anticipate increasing computing needs. Supercomputing: Is there anything you'd like to add? HARR: As cluster systems scale to thousands of processors, management becomes exponentially complex, and can be a daunting challenge for any organization. Therefore, cluster management is of vital importance as organizations must focus on their core competencies instead of spending valuable resources integrating, configuring and managing such systems. Linux NetworX has made it a priority to produce professional cluster management tools that are leading the cluster management industry. With the tools from Linux NetworX, organizations can spend more time developing their technology, not managing clusters. ---------- Supercomputing Online would like to thank Joshua Harr for his time and insights. ----------