HyperTransport sees bad economy driving HPC market
February 20, 2009 -- According to the HyperTransport Consortium, the need for cost efficiencies will drive trends toward virtualization and cloud computing, accelerating enterprise computing platform centralization.
February 20, 2009 -- Believing that the economic downturn makes computing technology key to reducing cost and increasing operational efficiency, the HyperTransport Technology Consortium says that it sees ongoing demand for optimized high-performance computing (HPC) infrastructure as capable of supporting job allocation, data handling and peak power flexibility.
Highlighting hardware and software virtualization and consolidation as vital enablers of cloud computing, consortium members discussed the applications and technologies that will be central to the high-performance computing market in the coming years at the International HyperTransport Symposium and Workshop 2009, held last week. The workshop took place at the HyperTransport Center of Excellence, which is managed by the Computer Architecture Group of the University of Heidelberg, in Mannheim, Germany.
"Demand for efficient data center and server farm infrastructure will continue to rise despite weaknesses in the global economy. Therefore, service provision players will need computing platforms that can support peak performance, real-time resource allocation and minimized Total Cost of Ownership (TCO)," commented Mario Cavalli, general manager of the HyperTransport Consortium. "Servers must offer an even more refined balance of performance, power consumption, mission-critical reliability and cost. As the industry's lowest latency, highest bandwidth and high-dependability interconnect technology, HyperTransport will be central to commercial enterprises' ability to meet this new HPC market challenge."
The consortium contends that cloud computing gives the commercial enterprise sector a way to outsource whole or part of its enterprise software application and customer relationship management (CRM) needs on a "pay-as-you-go" basis using Software as a Service (SaaS) techniques. The approach drastically reduces computing infrastructure and resource investments, says HyperTransport.
The consortium's recent releases of the HyperTransport 3.1 Link specification and the HTX3 HyperTransport Connector specification – supporting up to 51.2 GB/s and 20.8 GB/s of aggregate bandwidth respectively - enable the future-ready, high-performance computing infrastructure required to satisfy the industry's rising demand for performance, power and cost efficiency, as well as emerging HPC techniques like cloud computing.
The International HyperTransport Symposium and Workshop 2009 included speakers from both industry and academia, who provided updates on the most recent and upcoming developments in HyperTransport technology. The event explored the performance and TCO impact of HyperTransport on high-performance and fully scalable server cluster applications. Symposium presenters included:
"The mission of the HyperTransport Center of Excellence is to support and promote the use and applications of HyperTransport technology to research and industry, giving us the opportunity to highlight how these two groups have been working together so successfully in advancing the technology," concludes Cavalli. "With this year's Symposium, we have hosted the first International HyperTransport Technology Workshop, where members of academia showcased results of their HyperTransport-centric research projects and proposed new ideas for possible industry adoption."
On the Web: