A generator of numerically-tailored and high-throughput accelerators for batched GEMMs
Document typeConference report
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Rights accessOpen Access
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder
We propose a hardware generator of GEMM accelerators. Our generator produces vendor-agnostic HDL describing highly customizable systolic arrays guided by accuracy and energy efficiency goals. The generated arrays have three main novel aspects. First, the accelerators handle a large variety of computer number formats using intermediate representations based on our Sign Scale Significand (S3) format. Second, the processing elements perform all intermediate dot-product arithmetic operations required by the GEMM kernel without any intermediate rounding, which makes it possible to deliver better energy efficiency than state-of-the-art approaches while offering more accuracy and reproducible results. Third, our accelerators feature the Half-Speed Sink Down (HSSD) mechanism, which maximizes the overlap of host-accelerator data transfers with GEMM computations.We evaluate our automatically generated designs in a cutting-edge setup composed of a POWER9 host, CAPI (Coherent Accelerator Processor Interface) link, and a Virtex Ultrascale Plus FPGA. Arrays can operate at the speed of the link and saturate it to reach a 13GB/s throughput. Our fine-grain customization approach allows to cover a wide range of accuracy versus efficiency scenarios and can reach 0.65GOps/s/W while producing 1024 accurate bits or 148.7GOps/s/W with 6 accurate bits. Our configurations achieve up to 1613GOps/s system performance and power efficiencies of up to 240GOps/s/W for the FPGA. This automatic generator is the first being able to produce such a variety of designs. We improve the single-precision energy efficiency of state-of-the-art FPGA GEMM accelerators by 1.86×.
CitationLedoux, L.; Casas, M. A generator of numerically-tailored and high-throughput accelerators for batched GEMMs. A: IEEE Symposium on Field Programmable Custom Computing Machines. "2022 IEEE 30th International Symposium on Field-Programmable Custom Computing Machines, FCCM 2022: 15-18 May, 2022, New York, NY, USA: proceedings". Institute of Electrical and Electronics Engineers (IEEE), 2022, ISBN 978-1-6654-8332-2. DOI 10.1109/FCCM53951.2022.9786164.