Q/C Technologies, a San Francisco-based developer of optical computing systems, is working with a U.S. Department of Energy nanotechnology center to develop an optical processor that could eventually run artificial intelligence workloads using light rather than relying solely on electronic calculations.
The company announced a collaboration with Sandia National Laboratories’ Center for Integrated Nanotechnologies (CINT) to further develop its optical processing unit (OPU). The proposed hardware is primarily used for inference when trained models process new data to generate results.
Rather than performing calculations entirely through traditional electronic circuits, Q/C’s approach uses the interference of light to perform specific mathematical operations. Optical computing could potentially solve performance, bandwidth and scaling limitations associated with increasingly demanding computing workloads.
The project is still in the research stage. Q/C and CINT will study nanophotonic components and determine which optical operations and device technologies could form the building blocks of a scalable OPU architecture.
Light handles heavy computing tasks
A key goal is to create a technical roadmap to move beyond incremental improvements to graphics processing units. Researchers will examine some of the difficult problems that have prevented optical computers from becoming a mainstream alternative to electronic processors.
These challenges include performing nonlinear operations, integrating memory, maintaining computational accuracy, minimizing optical losses, and integrating photonic components into conventional electronics. Their solution will be important if individual optical components are to eventually work together as a practical computing system.
“Access to CINT’s scientific expertise and advanced research capabilities in photonics and nanotechnologies represents an important step in the development of our proprietary optical computing platform,” said Joshua Silverman, Chief Executive Officer of Q/C Technologies.
“As AI infrastructure faces increasing constraints on power, bandwidth and scalability, we believe optical computing has the potential to fundamentally transform the way compute-intensive AI workloads are processed.”
CINT gives the project access to expertise in nanoscience and nanophotonics at Sandia, one of the Department of Energy’s national laboratories. Sandia employs about 17,000 people and has an annual budget of about $5 billion. Research includes areas such as microelectronics, photonics, advanced materials and high-performance computing.
Building an optical processor
The collaboration will focus not only on individual photonic devices, but also on determining how these might ultimately be combined in Q/C’s proposed OPU. The company says the work should help identify and prioritize the technologies needed for its architecture.
“This collaboration provides an opportunity to accelerate our work on a practical, scalable optical computing architecture,” said Yossef Ehrlichman, chief technology officer at Q/C Technologies.
“We believe this work with CINT will help us define the path from basic optical capabilities to an integrated OPU architecture.”
According to Q/C, its longer-term goal is to develop optical processors with significantly higher clock speeds and bandwidths while consuming less energy than traditional computer architectures. These are development goals rather than proven performance figures for a finished processor.
The company is also building the infrastructure for the project. The company recently established a 50,000-square-foot integrated photonics laboratory in San Francisco to support its work in photonics, computing and OPU architecture development.
