Based on the insights from the evaluation, we propose a transparent, tractable and portable design framework for SFU-driven approximate acceleration on GPUs.

Jun 3, 2016 · Also, we exhaustively evaluate the transcendental functions that can be accelerated by SFUs in terms of their latency, throughput, accu- racy, ...

A transparent, tractable and portable design framework for SFU-driven approximate acceleration on GPUs is proposed that can provide fine-grained tuning for ...

Approximate computing, the technique that sacrifices certain amount of accuracy in exchange for substantial performance boost or power reduction, ...

To address this challenge, we focus on a very important but long ignored approximation unit on today's commercial GPUs | the specialfunction unit (SFU), and ...

SFU-Driven Transparent Approximation Acceleration on GPUs. A. Li, S. Song, M. Wijtvliet, A. Kumar, and H. Corporaal. ICS, page 15:1-15:14. ACM, (2016 ). 1.

Fingerprint. Dive into the research topics of 'SFU-driven transparent approximation acceleration on GPUs'. Together they form a unique fingerprint.

In this position paper, we introduce a transparent, tractable and portable design framework for SFU- driven approximate acceleration on GPUs, providing fine- ...

In this poster, we introduce a transparent, tractable, and portable design framework for SFU-driven approximate acceleration on GPUs, providing fine-grained ...

Ang Li, Shuaiwen Leon Song, Mark Wijtvliet, Akash Kumar, Henk Corporaal, "SFU-Driven Transparent Approximation Acceleration on GPUs", Proceedings of the 2016 ...