Today, Sun Microsystems’ Sun Fire servers were voted the 2005 SUPERCOMPUTING PRODUCT OF THE YEAR in Supercomputing Online's ANNUAL READER'S CHOICE AWARDS. The Awards identify significant technologies of the past year which promise to make the greatest impact on IT strategies and honor the products which exemplify the best implementation of those technologies. Looking at the results from our poll -- Sun’s Sun Fire servers win among the world’s leading products -- one would have to conclude that Sun Fire servers deliver the best technology to meet demanding application needs. We see many interesting products, but Sun's Sun Fire servers offer a unique combination of innovative technologies that can deliver exceptional performance, superior security and reliability. No longer limited to government scientists and academics, supercomputing or high-performance computer technology is being adopted by businesses of all sizes, across all industries. For example, auto makers rely on supercomputing technologies to simulate car crashes to make new cars safer. Energy companies leverage the power to find oil fields without having to drill a well. Financial Services organizations use it for portfolio management, stock market and economic forecasting. Did you know? The high-performance and technical computing market grew about 23% in 2005 to reach a record $9.1 billion in revenue. This is the third consecutive year of 20%-plus growth. The term "supercomputer" was coined by New York World newspaper in 1920. By using supercomputing applications, professionals are able to design, improve and get products to market faster because research and testing is done more effectively and efficiently. Plus, their utilization of computer resources is greatly increased, which results in improved ROI and reduced pressure on their computing budget. To learn more, we asked Bjorn Andersson, director of HPC and Grid Computing at Sun Microsystems, to share some reflections and projections. SC Online: How is supercomputing technology shaking up industry? How are enterprises using Sun’s Grid solutions in everyday business? Andersson: Fundamentally, supercomputing class installations are more and more implemented with Clusters and Grids of Clusters and other compute resources. This pushes down the threshold for access to this class of tools and makes it accessible over the network. It's really about today's incarnation of network computing with essentially unlimited scalability. Every company, almost regardless of the industry, now needs to view computer analysis and simulation as a business critical need, or they risk being out-innovated by companies, both smaller and larger. Supercomputing tools are becoming part of the mission critical path for the value a company delivers and how it can stay competitive in their market. SC Online: Please give us some details on how Sun Fire servers make supercomputing applications more accessible and maximize productivity (i.e. “time to insight”) for professionals. How does it compare to competitive offerings from other vendors? Andersson: Networked, scalable supercomputing is by its nature more affordable than the dedicated special purpose supercomputers of the past. So it's much more accessible from just an affordability aspect. The other trend that we're seeing is that as supercomputing moves into the mainstream, we're also seeing an increased need for enterprise class features. Supercomputing is becoming mission critical, so requirements for manageability and reliability are becoming much more important for these types of installation. Whereas in the past some system down time was acceptable, now redundancy in key components such as power supplies and fans are a more requested feature. Sun is also leveraging our experience from the enterprise class data centers and the processes we've built to be able to shorten the early phases of the life cycle, from architecting to deployment. We can deliver systems configured in racks that arrive ready to go at the customer site, enabling standardization for the customer and cutting out weeks of unproductive set-up time. The systems start to deliver real value right away. Essentially if you look at what Sun is doing, we are focusing our high performance computing (or supercomputing) offerings for real world problems. We're about providing the reliable, scalable tools for our customers to stay competitive. This means both having the fastest products and AMD Opteron processor-based x64 servers on the market, as well as being able to help the customer with the whole chain from configuring, architecting, deploying and managing. When you combine that with the fact that Sun has the longest experience in network computing, it really sets us apart from other supercomputing vendors. SC Online: Have you accomplished everything you sought to accomplish over the past year? What accomplishments are you most proud of? Andersson: We have been very busy across the line. Looking back over the past 12 months, we've been building out our family of world record setting Sun Fire x64 servers powered by AMD Opteron processors, establishing a new way of delivering high performance computing systems configured from the factory with our Sun Grid Rack System, and making key investments to reduce risk and time in high performance computing deployments for our customers. We invested in a Sun Solution Center for HPC in Hillsboro, Oregon. This is a customer-facing resource that is used to do proof of concept testing, sizing and performance benchmarking that helps our customers find the optimal configurations without having to experiment. The facility is today a 10 TeraFLOP capable cluster that already has played a key role in customer wins. One example is the supercomputer we're delivering to the Tokyo Institute of Technology. This will be the world's largest civilian supercomputer and is estimated to be in the Top 5 of the Top 500 supercomputer list. Wins like this are proving Sun's position as a leader in high performance computing. SC Online: Please let us in on why you think you’ve been so successful. Andersson: I'm convinced that our focus on solving real world problems really is what our customers are looking for. Based on our experience in enterprise class data centers and our roots in technical computing, we know the emerging requirements as high performance computing are becoming a required resource in many companies and the established IT processes are being applied. Sun is in a unique position in the market to be able to address both the traditional supercomputing requirements as well as the emerging needs, essentially providing the bridge to enterprise class supercomputing. Financial companies, the oil and gas industry, and academic institutions, are just a few examples of industries with compute-intensive applications that need quick, easy access to additional CPUs. We've responded with the introduction of the Sun Grid Compute Utility. Sun Grid helps customers and partners derive immediate benefits from an open, grid-based computing infrastructure on a utility basis by giving them more choice and control over how they purchase and leverage IT. By using a secure, standard multi-tenant computing, Sun is able to deliver a unique utility computing service that allows companies to share the cost of their computing with other corporations and moves the burden of complexity management onto the vendor. SC Online: How would you define your company’s strategy? Andersson: In terms of the supercomputing field, we're continually building a portfolio of products and solutions that address the changing requirements. You need to have your finger on the pulse of what's happening and also have access to the expertise that will take you to where you want to be. High performance computing has long been a leading-edge market and we see our investments in this market as being very strategic ones. Many products and technologies will make their way into regular IT infrastructure solutions down the road. Sun has long been a visionary leader in the computing industry and we have the people and partners that continue to push the limits. Sun's HPC business addresses a broad spectrum of customer need with three areas of focus -- HPC Grid, HPC Clusters, and Ultrascale, which is a term we use to distinguish the segment of customers who require the very highest levels of scalability and bandwidth. By leveraging Sun's strong ecosystems of channels, ISVs, and volume for cost reduction, and through our knowledge of the customer and ability to work closely with them in collaboration -- we can deliver the solution that best fits their need. Another piece to our HPC strategy is aggressive innovation. An example of this is our participation in DARPA's High Productivity Computing Systems (HPCS) project, where Sun is one of three companies selected to work on a next-generation high productivity solution to provide an order of magnitude breakthrough in time-to-solution. This is a multi-year commitment from Sun, and we have hundreds of people across the company involved in the project. Our HPCS leading-edge technology will make its way into our products and solutions at every level of our HPC business, as well as provide the foundation for our Ultrascale solutions. SC Online: Please tell us about Sun’s roadmap for future technology. Andersson: Of course, I can't talk about unannounced products, but we've publicly stated that we intend to continue to build out our Sun Fire x64 family and that we're committed to high performance computing as a strategic area for focus and growth. You'll see more scalability in the products and continued commitment to openness and choice for our customers. You have probably seen what we've done in terms of open sourcing and access to our software products over the last year, such as the OpenSolaris Operating System and providing free access to our development tools. These are very strategic moves that really resonate with many customers in the high performance computing space SC Online: Is there anything you’d like to add? Andersson: I'd like to stress that we're seeing a key shift in high performance computing -- it's expanding beyond its base in scientific R&D to also being used in more commercial settings. As supercomputing is becoming more mainstream, we also see enterprise class features being requested already today. Going forward we expect true scalability of application level performance to become more of a focus, as well as questions around security. In both cases, Sun is in an extremely good position with our experience in high-end SMP systems, in addition to the military-grade security that ships standard with the Solaris 10 OS. We're already seeing wins where the Solaris security functionality set is one of the key deciding factors. To top that off, we have the longest experience in the industry of providing scalable networked computing. Lastly, Sun's continuum of high performance computing offerings gives companies a choice between private or public grids, while also offering a variety of x64 servers and workstations, storage, software and services to meet any customer's computing needs. Supercomputing Online wishes to thank Bjorn Andersson for his time and insights.