FPGAs and GPUs ramp up to supercomputing

Reconfigurable, accelerating, scalable supercomputing uses a modular, interchangeable array of high-performance FPGAs and GPUs to deliver high performance, low latency and flexibility.

Available now from Sundance, the RASS reconfigurable, accelerating, scalable supercomputing solution has been created to easily integrate with existing high-performance computing (HPC) infrastructure and is supported by a widely available and mature design ecosystem.

Designed to integrate with industry standard processors and computing infrastructure, the RASS features a modular, interchangeable array of high-performance FPGAs and GPUs to deliver high performance, low latency and flexibility across single- and double-precision computation applications.

By exploiting parallel processing techniques and modular design, the RASS delivers a small-footprint energy-efficient high-density solution to applications in the capital markets, life sciences, national laboratories and computational fluid dynamics (CFD) At the heart of RASS are banks of tightly coupled Xilinx Virtex 5 FPGAs that include 640 Xilinx XtremeDSP slices capable of delivering up to 352GMACs of performance.

Sitting on a carrier, multiple FPGAs can be connected via a Rocket Serial Link (RSL) I/O interface and the architecture is adaptable to different applications and computation demands.

Over 20 different daughtercard and processing fabric combinations are available.

The SMT148-FX carrier offers SATA, USB 2.0, FireWire, Gigabit Ethernet, RS485, RS232, LVDS interfaces.

It also contains a Virtex-4 FX60 FPGA that includes two PowerPC cores that can be used as interface controllers, leaving the FPGAs free for extreme coprocessor acceleration.

"Closely coupling reprogrammable parallel processing architectures with other processing fabrics such as DSP and high-end GPU has been the core of our business for the last 20 years".

"By introducing RASS as a true platform technology, the dividend of this experience is now available on a scalable platform that can be tuned and easily adapted for high density computation across a range of applications", said Flemming Christensen, Managing Director of Sundance Multiprocessor Technology.

"With this single platform and portfolio of daughter cards the differentiation that RASS delivers is clear and our customers are optimising the computation mix more simply and effectively than ever before".

The massively parallel processing capabilities of RASS significantly lower energy consumption to yield lower cooling costs and a lower cooling BOM.

It delivers a market leading performance per watt and its small foot print enables significant real estate savings in space constrained data centres.

To mitigate the learning curve and risk of reliance on proprietary tools and languages, the RASS platform is supported by widely available industry standard design tools and languages.

Design support from the The MathWorks, Xilinx, 3L Diamond, Impulse, HDL and C/C++ providers offers a readily available and integrated development environment.

The design ecosystem enables immediate access to libraries of intellectual property including Navier-Stokes, Black-Scholes, Linpack, Monte Carlo and Amber, The RASS is immediately available from Sundance with prices subject to carrier and processing fabric configurations.

Back to top