Prism XL

SGI has introduced a new product for accelerator-based high-performance computing. Purpose-built to control the scale and speed of accelerators, SGI Prism XL enables users to scale to tens of Pflops or large Tflop range. The heterogeneous platform is built on the company’s new Stix architecture, which is based on a PCIe infrastructure for accelerator deployment, and can deliver up to a petaflop of computing power from a single cabinet, and a 20 petaflop double precision deployment in 100 cabinets.

Within the Stix architecture framework, Prism XL uses a 'stick' as the integral unit, with each stick containing two 'slices' of a full-length, full-height PCIe gen 2x 16 slot and a single-socket AMD 4100-powered motherboard. Each slice contains up to two 2.5 inch SATA drives and two 1.8 inch SSDs, which allows up to four Tbytes of storage on a single stick. It also contains built-in fans and auto-sensing power supplies to detect changes and keep the unit operating in almost any environment.

Supporting the latest accelerator cards from Nvidia, AMD and Tilera, and up to a 300W PCIe card, Stix architecture is accelerator-agnostic, providing users with flexibility to support the wave of accelerator and 'system on a chip' (SoC) options anticipated in the near future. Single-port or dual-port Mellanox ConnectX-2 cards are supported at first release, allowing up to two full bandwidth, fat-tree planes of 40 Gb/s InfiniBand.

Prism XL will also support CentOS 5.5 and RedHat RHEL 5.5. SGI Management Centre serves as the management layer, with Altair PBSpro GPU-ready for resource scheduling, and will allow users to tie together fully-capable hybrid environments containing Altix UV, Altix ICE, Rackable and Prism XL systems under a single management interface.


For functionality and security for externalised research, software providers have turned to the cloud, writes Sophia Ktori


Robert Roe looks at the latest simulation techniques used in the design of industrial and commercial vehicles


Robert Roe investigates the growth in cloud technology which is being driven by scientific, engineering and HPC workflows through application specific hardware


Robert Roe learns that the NASA advanced supercomputing division (NAS) is optimising energy efficiency and water usage to maximise the facility’s potential to deliver computing services to its user community


Robert Roe investigates the use of technologies in HPC that could help shape the design of future supercomputers