HPC Imaging

Matrox Supersight offers unprecedented performance and flexibility for machine vision and medical imaging applications that generate extraordinary amounts of data. Applications such as semiconductor wafer and mask inspection, or CT scanning benefit from performance by leveraging multiple clusters of CPUs, GPUs and FPGAs. Supersight provides an environment for considerable data and task-level parallel processing through the interconnection of CPUs, GPUs and FPGAs using a unique PCI-Express x16 2.0 (Gen2) switched fabric that removes I/O bottlenecks between the multiple processors.

Imaging applications differ significantly from traditional IT datacenter or HPC applications where the application is typically compute-bound. With imaging applications, the gigapixels/s of data leave most platforms I/O-bound. Supersight 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.

Supersight ensures that the performance gains from accelerators or additional processors are not lost during simple data transfers. The switched fabric lets developers recluster the compute elements to maintain optimal performance as the application is tuned or evolves with time.

Applications are developed with the Matrox MIL and its Distributed MIL API. Development on Supersight is simplified and ensures the source code is portable across all devices, from nodes with 32 CPU cores and 4 FPGAs, to a node with 8 CPU cores, 1 FPGA and 6 GPUs and everything in between.