ThUnderVolt: Enabling Aggressive Voltage Underscaling and Timing Error Resilience for Energy Efficient Deep Neural Network Accelerators
This addresses energy consumption issues for DNN accelerator users, offering a synergistic approach with existing techniques, though it is incremental in building on voltage underscaling methods.
The paper tackles the problem of energy inefficiency in DNN accelerators by proposing Thundervolt, a framework for aggressive voltage underscaling with timing error resilience, achieving 34%-57% energy savings on speech and image recognition benchmarks with less than 1% accuracy loss.
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.