Machine vision: New Matrox computing platform enables high-throughput image processing
Matrox Imaging announces its new high-performance computing (HPC) platform for industrial imaging available in third quarter 2009: Matrox Supersight e2. It enables 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 among multiple processors.
“Matrox Supersight e2 offers OEMs unprecedented performance and flexibility for machine vision and medical imaging applications that generate extraordinary amounts of data,” says Dwayne Crawford, product manager. Applications such as semiconductor wafer and mask inspection, flat panel display (FPD) inspection, and CT scanning will benefit from performance gains with Matrox Supersight e2 by leveraging multiple clusters of CPUs, GPUs and FPGAs, he says.
“We have seen tremendous improvements in imaging application performance with multiple CPU cores, pipelines in FPGAs, and GPUs as accelerators. But to get maximum performance, the technologies must be judiciously applied in a system,” says Crawford.
Imaging applications differ significantly from traditional IT datacenter or HPC applications where the application is typically compute-bound. With imaging applications, the gigapixels per second of data leave most platforms I/O-bound, says Crawford. “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.”
Applications for Matrox Supersight e2 are developed with the Matrox Imaging Library (MIL) and its Distributed MIL API. As MIL is supported on all Matrox Imaging hardware platforms, development on Matrox Supersight e2 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.
With up to 7 years of availability for Matrox products , Crawford contends OEMs get the maximum return on investment by avoiding costly and constant re-validation of ever-changing commercial PC platforms. “Though the nature of the computing industry makes change unavoidable, Matrox Supersight e2, like all Matrox Imaging hardware, benefits from rigorous managed life cycle programs that range from standard long-term fit, form and function compatibility to strict revision control, including prior-change notification, risk analysis and acceptance sign-off,” he says.
Vision system company Matrox Imaging launches integrator program .
More Machine Control and Machine Vision news from Control Engineering
– Edited by Renee Robbins, senior editor
Control Engineering News Desk