Thundervolt: enabling aggressive voltage underscaling and timing error resilience for energy efficient deep learning accelerators
Proceedings of the 55th Annual Design Automation Conference, 2018•dl.acm.org
Hardware accelerators are being increasingly deployed to boost the performance and
energy efficiency of deep neural network (DNN) inference. In this paper we propose
Thundervolt, a new framework that enables aggressive voltage underscaling of high-
performance DNN accelerators without compromising classification accuracy even in the
presence of high timing error rates. Using post-synthesis timing simulations of a DNN
accelerator modeled on the Google TPU, we show that Thundervolt enables between 34 …
energy efficiency of deep neural network (DNN) inference. In this paper we propose
Thundervolt, a new framework that enables aggressive voltage underscaling of high-
performance DNN accelerators without compromising classification accuracy even in the
presence of high timing error rates. Using post-synthesis timing simulations of a DNN
accelerator modeled on the Google TPU, we show that Thundervolt enables between 34 …
Hardware accelerators are being increasingly deployed to boost the performance and energy efficiency of deep neural network (DNN) inference. In this paper we propose Thundervolt, a new framework that enables aggressive voltage underscaling of high-performance DNN accelerators without compromising classification accuracy even in the presence of high timing error rates. Using post-synthesis timing simulations of a DNN accelerator modeled on the Google TPU, we show that Thundervolt enables between 34%-57% energy savings on state-of-the-art speech and image recognition benchmarks with less than 1% loss in classification accuracy and no performance loss. Further, we show that Thundervolt is synergistic with and can further increase the energy efficiency of commonly used run-time DNN pruning techniques like Zero-Skip.
ACM Digital Library