GATE: A Generalized Dataflow-level Approximation Tuning Engine For Data Parallel Architectures
- Title
- GATE: A Generalized Dataflow-level Approximation Tuning Engine For Data Parallel Architectures
- Author
- 박영준
- Issue Date
- 2019-06
- Publisher
- Association for Computing Machinery
- Citation
- DAC '19: Proceedings of the 56th Annual Design Automation Conference 2019, Page. 1-6
- Abstract
- Although approximate computing is widely used, it requires substantial programming effort to find appropriate approximation patterns among multiple pre-defined patterns to achieve a high performance. Therefore, we propose an automatic approximation framework called GATE to uncover hidden opportunities from any data-parallel program regardless of the code pattern or application characteristics using two compiler techniques, namely subgraph-level approximation (SGLA) and approximate thread merge(ATM). GATE also features conservative/aggressive tuning and dynamic calibration to maximize the performance while maintaining the TOQ level during runtime. Our framework achieves an average performance gain of 2.54x over the baseline with minimum accuracy loss.
- URI
- https://dl.acm.org/doi/10.1145/3316781.3317833https://repository.hanyang.ac.kr/handle/20.500.11754/151854
- ISBN
- 9781450367257
- ISSN
- 0738-100X
- DOI
- 10.1145/3316781.3317833
- Appears in Collections:
- COLLEGE OF ENGINEERING[S](공과대학) > COMPUTER SCIENCE(컴퓨터소프트웨어학부) > Articles
- Files in This Item:
There are no files associated with this item.
- Export
- RIS (EndNote)
- XLS (Excel)
- XML