AMD Opteron 6100 Series Processor to provide a New Option for HPC Innovation

Appro today announced that it will support AMD's upcoming server platform, currently codenamed "Maranello," which is scheduled for availability in Q1 2010.  AMD's upcoming server platform is expected to fit within the new G34 socket and feature the eight- and twelve-core AMD Opteron 6100 Series processor variant currently codenamed "Magny-Cours". Appro plans to showcase a technology sample of this new platform at the Supercomputing show in Portland, OR, booth # 1200, November 17-19, 2009.

Appro expects to refresh its dual and quad-socket performance optimized and power-efficient supercomputing building block server lines based on the next-generation "Maranello" platform. Appro expects that its Hyper Cluster Series based on these new server platforms will offer the latest optimized technologies for HPC such as DDR3, quad 16-bit HyperTransport 3 technology links with up to 6.4 GT/s per link, and the AMD SR56x0 chipset with AMD-Vi technology and PCIe Gen 2.

"Appro is proud to collaborate with AMD for the upcoming AMD Opteron 6100 Series processor-based server clusters solutions," John Lee, VP of Advanced Technology Solutions of Appro. "Working together, Appro and AMD provide industry-leading performance-per-watt capabilities to meet the demands of our High-Performance computing customers."

"AMD and Appro have a commitment to delivering HPC solutions to customers that address both performance and energy efficiency, while continuing to push the envelope on scalability for critical workloads," said Patrick Patla, vice president and general manager, Server/Workstation and Embedded division, AMD. "AMD's next-generation server technology is not only expected to deliver outstanding eight- and twelve-core performance, but also to adapt to evolving market demands. New enhancements, in addition to the inherent architectural advantages of the AMD Opteron processor, deliver advanced memory bandwidth and performance-per-watt, and allow customers to tackle important high performance computing applications with efficiency."