SSAGA
| Autor: | Sanghamitra Roy, Prabal Basu, Chidhambaranathan Rajamanikkam, Shamik Saha, Koushik Chakraborty, Aatreyi Bal |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
010302 applied physics
High energy Power gating Computer science CAD Multiprocessing 02 engineering and technology Parallel computing 01 natural sciences Computer Graphics and Computer-Aided Design 020202 computer hardware & architecture Computer Science Applications 0103 physical sciences 0202 electrical engineering electronic engineering information engineering Human multitasking Electrical and Electronic Engineering General-purpose computing on graphics processing units Design paradigm Efficient energy use |
| Zdroj: | ACM Transactions on Design Automation of Electronic Systems. 22:1-20 |
| ISSN: | 1557-7309 1084-4309 |
| DOI: | 10.1145/3014163 |
| Popis: | The emergence of GPGPU applications, bolstered by flexible GPU programming platforms, has created a tremendous challenge in maintaining high energy efficiency in modern GPUs. In this article, we demonstrate that customizing a Streaming Multiprocessor (SM) of a GPU at a lower frequency is significantly more energy efficient compared to employing DVFS on an SM designed for a high-frequency operation. Using a system-level CAD technique, we propose SSAGA—Streaming Multiprocessors Synthesized for Asymmetric GPGPU Applications —an energy-efficient GPU design paradigm. SSAGA creates architecturally identical SM cores, customized for different voltage-frequency domains. Our rigorous cross-layer methodology demonstrates an average of 20% improvement in energy efficiency over a spatially multitasking GPU across a range of GPGPU applications. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|