Dynamic Slack-Aware Clocking for Near-Threshold Tensor Processing Units (TPUs)
Open Access
- 17 August 2026
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Design Automation of Electronic Systems
Abstract
Operating Tensor Processing Units (TPUs) in the near-threshold computing (NTC) region significantly reduces energy consumption but introduces high delay sensitivity to process variation and data activity. Conventional designs typically rely on a conservative, fixed global clock to ensure safety, which leaves large portions of timing margin unexploited as most operations finish well before the clock edge. We propose Dynamic Slack-Aware Clocking (DSAC), a proactive framework that replaces worst-case timing with operation-specific adjustments. DSAC employs lightweight Hamming-Distance, Most-Significant-Bit, and Hybrid predictors to estimate the delay sensitivity of individual multiply-accumulate (MAC) operations and classify them into three timing tiers. These tiers are enforced locally via dummy-hold cycles under a fixed global reference clock, enabling fine-grained timing adaptation without global clock retuning or frequency scaling. A closed-loop feedback controller monitors timing violations and updates tier thresholds at runtime to maintain resilience. Experiments on quantized DNN benchmarks demonstrate that the MSB predictor maintains high inference accuracy, with an average loss of only 1% even at aggressive performance points. Furthermore, DSAC achieves up to 1.55 × better energy efficiency at 2.15 × frequency scaling compared to a baseline TPU, while incurring an area overhead as low as 13%.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- SCISSORS: System Level Error Detection for Enabling Near-Threshold Operating Systolic ArraysIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2025
- ThUnderVolt: Enabling Aggressive Voltage Underscaling and Timing Error Resilience for Energy Efficient Deep Learning Accelerators2018
- Motivation for and Evaluation of the First Tensor Processing UnitIEEE Micro, 2018
- VARIUS-TC: A modular architecture-level model of parametric variation for thin-channel switches2016
- Deep Residual Learning for Image Recognition2016
- Man vs. computer: Benchmarking machine learning algorithms for traffic sign recognitionNeural Networks, 2012
- VARIUS-NTV: A microarchitectural model to capture the increased sensitivity of manycores to process variations at near-threshold voltages2012
- Near-Threshold Computing: Reclaiming Moore's Law Through Energy Efficient Integrated CircuitsProceedings of the IEEE, 2010
- GPU ComputingProceedings of the IEEE, 2008
- New Generation of Predictive Technology Model for Sub-45 nm Early Design ExplorationIEEE Transactions on Electron Devices, 2006