Modeling Deep Learning Accelerator Enabled GPUs
| Autor: | Aamir Raihan, Tor M. Aamodt, Negar Goli |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
010302 applied physics
Hardware architecture Instructions per cycle business.industry Computer science Deep learning 02 engineering and technology Software_PROGRAMMINGTECHNIQUES 01 natural sciences 020202 computer hardware & architecture Computational science CUDA Titan (supercomputer) Template 0103 physical sciences 0202 electrical engineering electronic engineering information engineering Artificial intelligence Core model business Architectural model ComputingMethodologies_COMPUTERGRAPHICS |
| Zdroj: | ISPASS |
| DOI: | 10.1109/ispass.2019.00016 |
| Popis: | The efficacy of deep learning has resulted in its use in a growing number of applications. The Volta graphics processor unit (GPU) architecture from NVIDIA introduced a specialized functional unit, the “tensor core”, that helps meet the growing demand for higher performance for deep learning. In this paper we study the design of the tensor cores in NVIDIA's Volta and Turing architectures. We further propose an architectural model for the tensor cores in Volta. When implemented a GPU simulator, GPGPU-Sim, our tensor core model achieves 99.6% correlation versus an NVIDIA Titan V GPU in terms of average instructions per cycle when running tensor core enabled GEMM workloads. We also describe support added to enable GPGPU-Sim to run CUTLASS, an open-source CUDA C++ template library providing customizable GEMM templates that utilize tensor cores. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|