Saving Energy with Per-Variable Bitwidth Speculation
Tommy McMichen, David Dlott, Panitan Wongse-ammat, Nathan Greiner, Hussain Khajanchi, Russ Joseph, Simone Campanoni
Abstract
Tiny devices have become ubiquitous in people's daily lives. Their applications dictate tight energy budgets, but also require reasonable performance to meet user expectations. To this end, the hardware of tiny devices has been highly optimized, making further optimizations difficult. In this work, we identify a missed opportunity: the bitwidth selection of program variables. Today's compilers directly translate the bitwidth specified in the source code to the binary. However, we observe that most variables do not utilize the full bitwidth specified in the source code for the majority of execution. To leverage this opportunity, we propose BitSpec : a system that performs fine-grained speculation on the bitwidth of program variables. BitSpec is implemented as a compiler-architecture co-design, where the compiler transparently reduces the bitwidth of program variables to their expected needs and the hardware monitors speculative variables, reporting misspeculation to the software, which re-executes at the original bitwidth, ensuring correctness. BitSpec reduces energy consumption by 9.9% on average, up to 28.2% .
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Builds on2
- DSAGEN: Synthesizing Programmable Spatial AcceleratorsJian Weng, Sihao Liu, Vidushi Dadu, Zhengrong Wang et al.ISCA 2020 · 140 citations
- Snafu: An Ultra-Low-Power, Energy-Minimal CGRA-Generation Framework and ArchitectureGraham Gobieski, Ahmet Oguz Atli, Kenneth Mai, Brandon Lucia et al.ISCA 2021 · 84 citations
Related papers
- Challenging Sequential Bitstream Processing via Principled Bitwise SpeculationJunqiao Qiu, Lin Jiang, Zhijia ZhaoASPLOS 2020 · 9 citations
- Enabling Floating Point Virtualization With Tiny NumbersKevin Hayes, Peter A. DindaHPDC 2026
- A Model-Specific End-to-End Design Methodology for Resource-Constrained TinyML HardwareYanchi Dong, Tianyu Jia, Kaixuan Du, Yiqi Jing et al.DAC 2023 · 10 citations
- Zero-Value Code Specialization via Profile-Guided Control Data Flow AnalysisShaokang Du, Kelun Lei, Xin You, Hailong Yang et al.SC 2025 · 1 citation
- T4: Compiling Sequential Code for Effective Speculative Parallelization in HardwareVictor A. Ying, Mark C. Jeffrey, Daniel SánchezISCA 2020 · 25 citations
