SCISSORS: System Level Error Detection for Enabling Near-Threshold Operating Systolic Arrays
- 1 July 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Vol. 45 (2) , 719-731
Abstract
Since dynamic power has a quadratic relationship with voltage, reducing voltage is an effective way to lower power consumption in digital circuits. However, maintaining stable operation at lower voltages is challenging due to increased sensitivity to process, voltage, and temperature (PVT) variations, making it difficult to determine optimal operating points using static timing analysis (STA). While circuit- and device-level solutions like timing error detection (TED) systems can enable lower voltage operation, they introduce significant overhead and design complexity. In this article, we integrate an algorithm-based fault tolerance (ABFT) method into the structure of systolic arrays (SAs) to capture timing errors when voltage is scaled down, ensuring safe and optimized low-voltage operation. Our proposed approach, SCISSORS, demonstrates how extra voltage margins in SAs used for matrix arithmetic can be trimmed by integrating a simple algorithmic technique into the structure of the array. This solution not only detects errors in the accelerator but also those caused by voltage reduction in on-chip memory and auxiliary circuits. It is fully implementable through hardware description languages (HDL) without requiring transistor- or circuit-level modifications to the netlist. Implementation on a Zynq system-on-chip (SoC) shows that SCISSORS introduces only a tolerable overhead of 11% and 8% for $32{\times }32$ and $64{\times }64$ SAs, respectively, while achieving nearly a $2{\times }$ improvement in energy efficiency. Experimental results further demonstrate that SCISSORS adaptively adjusts voltage in response to the voltage-temperature coupling behavior of digital circuits at runtime, specifically addressing inverse temperature dependence (ITD).Keywords
Funding Information
- German Research Foundation (“Deutsche Forschungsgemeinschaft”) (287022738 TRR 196)
- German Research Foundation (“Deutsche Forschungsgemeinschaft”) (S05)
- Research Council of Finland through 6G Flagship (369116)
This publication has 41 references indexed in Scilit:
- Error Checking for Sparse Systolic Tensor Arrays2024
- RunSAFER: A Novel Runtime Fault Detection Approach for Systolic Array Accelerators2023
- A Comprehensive Analysis of Transient Errors on Systolic Arrays2023
- Modeling TPU Thermal Maps Under Superlattice Thermoelectric CoolingIEEE Access, 2022
- A 0.4-0.9V, 2.87pJ/cycle Near-Threshold ARM Cortex-M3 CPU with In-Situ Monitoring and Adaptive-Logic Scan2020
- RISC-V Based MPSoC Design Exploration for FPGAs: Area, Power and Performance2020
- Modern Hardware Margins: CPUs, GPUs, FPGAs Recent System-Level Studies2019
- ThUnderVolt: Enabling Aggressive Voltage Underscaling and Timing Error Resilience for Energy Efficient Deep Learning Accelerators2018
- ADMM-based infinity norm detection for large MU-MIMO: Algorithm and VLSI architecture2017
- Measure twice and cut once: Robust dynamic voltage scaling for FPGAs2016