000341824 001__ 341824
000341824 005__ 20230827173327.0
000341824 0247_ $$2CORDIS$$aG:(EU-Grant)101113120$$d101113120
000341824 0247_ $$2CORDIS$$aG:(EU-Call)ERC-2022-POC2$$dERC-2022-POC2
000341824 0247_ $$2originalID$$acorda_____he::101113120
000341824 035__ $$aG:(EU-Grant)101113120
000341824 150__ $$aFast Matrix Multiplication for AI$$y2023-04-01 - 2024-09-30
000341824 372__ $$aERC-2022-POC2$$s2023-04-01$$t2024-09-30
000341824 450__ $$aFMMF-AI$$wd$$y2023-04-01 - 2024-09-30
000341824 5101_ $$0I:(DE-588b)5098525-5$$2CORDIS$$aEuropean Union
000341824 680__ $$aMatrix multiplication consumes huge amount of resources: computing time and energy, primarily in AI applications. The industry has recognized the need for faster and more energy-efficient matrix multiplication with state-of-the-art solutions in software (e.g., DGEMM of Intel's math kernel library (MKL) for CPU and NVIDIA's CUDA for GPU) and hardware (e.g., Google's TPU and Intel / Habana labs Gaudi accelerator). Unfortunately, all present solutions employ a wasteful cubic-time algorithm. We have developed methods that provide speedup for matrix multiplication in SW and in HW. The novel developments of Prof. Oded Schwartz and his strong team are based on years of research, and are protected by several patents. The funds are requested to pursue business opportunity.
000341824 909CO $$ooai:juser.fz-juelich.de:1011183$$pauthority:GRANT$$pauthority
000341824 909CO $$ooai:juser.fz-juelich.de:1011183
000341824 980__ $$aG
000341824 980__ $$aCORDIS
000341824 980__ $$aAUTHORITY