June 21, 2017 -- MSU News Service
S2 Corporation, of Bozeman, and Montana State University’s Spectrum Lab, announce the award of a $1 million dollar, 12-month research contract from the Intelligence Advanced Research Projects Agency, IARPA, to develop an efficient, high-data rate photonic computational engine for 2-D image processing. Applications include virus detection in streaming digital data, key features search for computer vision, and queries in massive unindexed databases.
The S2 processor-in-memory capability is analog and photonic based, using laser light to interact with a crystal, and falls within a new class of technology different from typical integrated circuits. Microprocessors, microcontrollers, memory, and other digital logic circuits use complementary metal–oxide–semiconductor, CMOS, transistor architectures. The S2 approach is of interest to IARPA’s Filtering and Selection Technology, FaST, program for its extremely high data rate and performance in wall power efficiency compared to the typical CMOS based approaches.
The new work builds on prior IARPA-funded efforts at S2 over the past two years, which propelled early S2 capability for real-time streaming data search by 10,000-fold; up to high single line rates of 200 gigabits per second and, very importantly, with a simultaneous 100-fold reduction in power relative to state-of-the-art digital supercomputers. The prior work, soon to be published in Applied Optics, established plausibility of real-time, key feature identification in streaming data and large unindexed databases, and circumnavigates the otherwise insurmountable memory latency delays and associated energy cost penalties when using conventional CMOS processors. The prior work completed in 2016 also demonstrated the plausible potential for further scaling of S2 capability to an extraordinary 10 terabits per second throughput for data processing, by advanced engineering efforts.
The new work, started in March 2017, includes a subcontract to Montana State University.
The S2 team is pursuing this area to address the global challenges of processing increasingly massive data in real time and with energy efficiency.
“The interest is there,” said Kris Merkel, CEO of S2 Corporation, “and we are working to advance the power of S2 technology to meet strategic global needs that currently outpace the capability of conventional electronics.”
Zeb Barber, director of Spectrum Lab at MSU, said, “The fundamental research at MSU allows opportunities for faculty and students to advance the state-of-art in new realms of physics-based algorithms and applications, and to work on cutting-edge technology alongside industry.”
IARPA recognized the potential of S2 technology several years ago, and encouraged this new direction of research.
“S2 technology provides a near instantaneous Fourier transform of streaming data, storing the vast spectral components as microscopic holograms inside a cold crystal, and permitting real-time multiplication of data at unprecedented clock rates and power-efficiency. Dot-product-engines of this kind possess immense significance to science and engineering. If successful, and the challenge remains high, it is difficult to overstate the significance of a potential breakthrough in streaming 2D imagery analysis for computer vision,” said Karl Roenigk, IARPA program manager.
Contact: Kris Merkel, president and CEO, S2 Corporation, (406) 922-0334, merkel@s2corporation.com; Zeb Barber, director of MSU's Spectrum Lab, (406) 994-5925, barber@spectrum.montana.edu