Matrox Imaging has announced a new high performance computing platform designed to tackle industrial imaging applications. The Matrox Supersight e2 allows one to interconnect CPUs, GPUs and FPGAs through the use of a unique PCI Express x16 [16 lane Gen2.0] switched fabric. According to product manager Dwayne Crawford, imaging applications leave most traditional compute platforms lacking in performance due to their unusual I/O-bound nature.
Matrox Supersight e2 offers a solution to this unique architectural challenge as traditional Blade servers and 1U ‘pizza boxes’ simply do not have the I/O bandwidth to distribute images to the processors and accelerators,” he says. “Matrox Supersight e2 ensures that the performance gains from accelerators or additional processors are not lost during simple data transfers, as seen in Blade or 1U pizza box environments. Furthermore, the switched fabric lets developers “re-cluster” the compute elements to maintain optimal performance as the application is tuned or evolves with time.
This is a very interesting approach to solving, what has become, a real issue in today’s HPC market. More and more users are beginning to adopt non-traditional computing methods [such as GPUs]. Many have quickly realized that selecting a suitable host platform for hosting more than a single accelerator is quite difficult.
For more info on the new digs at Matrox, read the full release here.