“Data transfer between core and memory has become the limiting factor in computational speed. AI and Big Data applications need to deal with large data sets, which only intensifies the need for a next generation architecture,” said Jay Kramer, Chairman of the Awards Program, “we are proud to recognize Blueshift Memory for the innovation of Cambridge Architecture. This technology provides a new level of computational efficiency with zero latency and up to 1000 times faster memory access, while providing up to a 50% reduction in energy consumption by eliminating unnecessary movement of data.”
The demonstration device was reported to have achieved an acceleration factor of 16 to 128 times for processing image data, along with ultra-low power consumption.
Work continues on realising the potential of the architecture in big data applications, and on applying quantum resilient data protection to the chip design.