Startup Runs AI in Novel SRAM
Areanna claims 100 TOPS/W in simulation
By Rick Merritt, EETimes
July 22, 2019
SAN JOSE — In his spare time, an engineer at Tektronix sketched out a novel deep-learning accelerator, and now his two-person startup is the latest example of the groundswell of enthusiasm that deep learning is generating.
Behdad Youssefi defined an SRAM with specialized cells that can handle the matrix multiplication, quantization, storage and other jobs needed for an inference processor. After four years solo work on the concept originally planned as a PhD thesis, he formed startup Areanna with a colleague at Tektronix and a Berkeley professor as an advisor.
In Spice simulations the design delvers more than 100 tera-operations/second/watt when recognizing handwritten digits using 8-bit integer math. Youssefi claims it could beat Google’s TPU in computational density by an order of magnitude.
E-mail This Article | Printer-Friendly Page |
Related News
- Israeli AI startup NeuReality raises $35M Series A to bring its novel inferencing chip to the market
- Dream Chip Technologies tapes out a 10-TOPS SoC in 22nm with a novel AI Accelerator and an Automotive Functional Safety Processor
- SEMIFIVE announces commercialization of its 5nm HPC SoC Platform with lead partner Rebellions, AI Chipmaker startup based in Korea
- Andes Technology Announces The New Product Line, AndesAIRE™, Ultimately Efficient AI/ML Solutions For Edge And End-Point Inference
- Networking Chip Startup Enfabrica Emerges from Stealth Mode to Solve Scalability and Price-Performance Challenges for AI Growth in Cloud
Breaking News
- Packetcraft Delivers Bluetooth Software Enabling Channel Sounding to Early Access Licensees
- QuickLogic Releases Aurora 2.6 with Expanded Operating System Support and Up to 15% Faster Performance
- Rapid Silicon Introduces Revolutionary Rapid eFPGA Configurator for Hassle Free Embedded FPGA Evaluation
- Silvaco Announces Launch of Initial Public Offering
- Rambus Reports First Quarter 2024 Financial Results