Mercury releases 10 Teraflop ISR Subsystem Capability
CHELMSFORD, Mass., 12 May 2012. Mercury Computer Systems, Inc. (NASDAQ: MRCY), a provider of ISR and EW subsystems based in Chelmsford, Mass, has released StreamDirect, a method for delivering streams of sensor data to specialized coprocessors such as general purpose graphics processing units (GPGPUs).
CHELMSFORD, Mass., 12 May 2012. Mercury Computer Systems, Inc. (NASDAQ: MRCY), a provider of ISR and EW subsystems based in Chelmsford, Mass., has released StreamDirect, a method for delivering streams of sensor data to specialized coprocessors such as general purpose graphics processing units (GPGPUs).
StreamDirect, designed for use in areas such as military embedded systems, enables transfers from the I/O sensors directly to the GPGPUs, supporting over 10 teraflops (TFLOPS) of processing capability in a rugged OpenVPX system. StreamDirect has been deployed in previous Mercury products, but it is now being made available in Mercury's standard product, the 6U OpenVPX GSC6201. The GSC6201 is a carrier card that uses industry-standard GPGPU MXMs and is designed to accept those based on the NVIDIA Fermi and Kepler architectures.
StreamDirect enables direct communication of data from the source, such as a sensor input device, into a coprocessor's memory, such as a GPGPU, without intermediate storage in the CPU. Most systems that use GPGPUs have to first pass the data to a CPU's memory and then transfer the same data from the CPU's memory to the GPGPU. StreamDirect eliminates the copy step, creating a high-bandwidth DMA channel between the sensor and the GPGPU. StreamDirect uses Mercury's POET/ICS technology and NVIDIA GPUDirect to provide a system-wide communication capability that enables applications such as EO/IR, radar, cyber and electronic warfare to benefit from quicker data delivery.
To keep up with the advances in GPGPU innovations, Mercury is using the MXM form factor to deliver GPGPU technology in a rugged OpenVPX module. The GSC6201 can be upgraded to other GPGPU architectures, such as the embedded Kepler MXM, when they are available.
NVIDIA CUDA, OpenCL and Mercury's Scientific Algorithm Library (SAL) software development platforms are supported on the GSC6201, offering processing functions based on open standards, cross-platform support, and porting of legacy applications to leverage parallel stream processing with GPGPUs.
The GSC6201 is available in commercial and rugged versions including air- and conduction-cooled configurations. Mercury offers StreamDirect for NVIDIA GPGPUs as well.
For more information contact Mercury online at www.mc.com