research-article

TAFFO: Tuning Assistant for Floating to Fixed Point Optimization

Authors:

Stefano Cherubin,

Daniele Cattaneo,

Michele Chiari,

Antonio Di Bello,

Giovanni AgostaAuthors Info & Claims

IEEE Embedded Systems Letters, Volume 12, Issue 1

Pages 5 - 8

https://doi.org/10.1109/LES.2019.2913774

Published: 01 March 2020 Publication History

Abstract

While many approximate computing methods are quite application-dependent, reducing the size of the data representation used in the computation has a more general applicability. We present a tuning assistant for floating to fixed point optimization (TAFFO), an LLVM-based framework designed to assist programmers in the precision tuning of software. We discuss the framework architecture and we provide guidelines to effectively tradeoff precision to improve the time-to-solution. We evaluate our framework on a well-known approximate computing benchmark suite, AxBench, achieving a speedup on 5 out of 6 benchmarks (up to 366%) with only a limited loss in precision (<3% for all benchmarks). Contrary to most related tools, TAFFO supports both C and C++ programs. It is provided as a plugin for LLVM, a design solution that improves significantly the maintainability of the tool and its ease of use.

References

[1]

L. Breiman, “Bagging predictors,” Mach. Learn., vol. 24, no. 2, pp. 123–140, Aug. 1996.

[2]

D. Cattaneo, A. D. Bello, S. Cherubin, F. Terraneo, and G. Agosta, “Embedded operating system optimization through floating to fixed point compiler transformation,” in Proc. 21st Euromicro Conf. Digit. Syst. Design (DSD), 2018, pp. 172–176.

[3]

E. Darulova and V. Kuncak, “Trustworthy numerical computation in scala,” SIGPLAN Notices, vol. 46, no. 10, pp. 325–344, Oct. 2011.

Digital Library

[4]

E. Darulova and V. Kuncak, “Towards a compiler for reals,” ACM Trans. Program. Lang. Syst., vol. 39, no. 2, p. 8, Mar. 2017.

[5]

L. H. de Figueiredo and J. Stolfi, “Affine arithmetic: Concepts and applications,” Numer. Algorithms, vol. 37, nos. 1–4, pp. 147–158, Dec. 2004.

[6]

IEEE Computer Society Standards Committee. Floating-Point Working Group of the Microprocessor Standards Subcommittee, IEEE Standard for Floating-Point Arithmetic, IEEE Standard, pp. 1–70, Aug. 2008.

[7]

H. Keding, M. Willems, M. Coors, and H. Meyr, “FRIDGE: A fixed-point design and simulation environment,” in Proc. Conf. Design Autom. Test Europe, Paris, France, 1998, pp. 429–435.

[8]

K.-I. Kum, J. Kang, and W. Sung, “AUTOSCALER for C: An optimizing floating-point to integer C program converter for fixed-point digital signal processors,” IEEE Trans. Circuits Syst. II, Analog Digit. Signal Process., vol. 47, no. 9, pp. 840–848, Sep. 2000.

[9]

M. O. Lam, J. K. Hollingsworth, B. R. de Supinski, and M. P. Legendre, “Automatically adapting programs for mixed-precision floating-point computation,” in Proc. 27th Int. ACM Conf. Supercomput. (ICS), Eugene, OR, USA, 2013, pp. 369–378.

[10]

C. Lattner and V. Adve, “LLVM: A compilation framework for lifelong program analysis & transformation,” in Proc. Int. Symp. Code Gener. Optim., San Jose, CA, USA, 2004, pp. 75–86.

[11]

S. Mitra, M. K. Gupta, S. Misailovic, and S. Bagchi, “Phase-aware optimization in approximate computing,” in Proc. Int. Symp. Code Gener. Optim., Austin, TX, USA, 2017, pp. 185–196.

[12]

R. E. Mooreet al., Introduction to Interval Analysis. Philadelphia, PA, USA: SIAM, 2009.

[13]

F. Pedregosaet al., “Scikit-learn: Machine learning in python,” J. Mach. Learn. Res., vol. 12, pp. 2825–2830, Oct. 2011.

Digital Library

[14]

C. Rubio-Gonzálezet al., “Precimonious: Tuning assistant for floating-point precision,” in Proc. Int. Conf. High Perform. Comput. Network. Storage Anal. (SC), Denver, CO, USA, Nov. 2013, pp. 1–12.

[15]

C. Rubio-Gonzálezet al., “Floating-point precision tuning using blame analysis,” in Proc. 38th Int. Conf. Softw. Eng. (ICSE), Austin, TX, USA, 2016, pp. 1074–1085.

[16]

A. Sampsonet al., “ACCEPT: A programmer-guided compiler framework for practical approximate computing,” Univ. Washington, Seattle, WA, USA, Rep. UW-CSE-15-01-01, Apr. 2015.

[17]

A. Yazdanbakhsh, D. Mahajan, H. Esmaeilzadeh, and P. Lotfi-Kamran, “AxBench: A multiplatform benchmark suite for approximate computing,” IEEE Des. Test., vol. 34, no. 2, pp. 60–68, Apr. 2017.

[18]

T. Yuki, “Understanding PolyBench/C 3.2 kernels,” in Proc. Int. Workshop Polyhedral Compilation Tech. (IMPACT), 2014, pp. 1–5.

[19]

Q. Zhang, F. Yuan, R. Ye, and Q. Xu, “Approxit: An approximate computing framework for iterative methods,” in Proc. Design Autom. Conf. (DAC), San Francisco, CA, USA, Jun. 2014, pp. 1–6.

Cited By

Xu JSong GZhou BLi FHao JZhao JLee IChabbi MSteuwer M(2024)A Holistic Approach to Automatic Mixed-Precision Code Generation and Tuning for Affine ProgramsProceedings of the 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming10.1145/3627535.3638484(55-67)Online publication date: 2-Mar-2024
https://dl.acm.org/doi/10.1145/3627535.3638484
Cattaneo DMaggioli AMagnani GDenisov LYang SAgosta GCherubin S(2023)Mixed Precision in Heterogeneous Parallel Computing Platforms via Delayed Code AnalysisEmbedded Computer Systems: Architectures, Modeling, and Simulation10.1007/978-3-031-46077-7_33(469-477)Online publication date: 2-Jul-2023
https://dl.acm.org/doi/10.1007/978-3-031-46077-7_33
Zoni DGalimberti A(2022)Cost-effective fixed-point hardware support for RISC-V embedded systemsJournal of Systems Architecture: the EUROMICRO Journal10.1016/j.sysarc.2022.102476126:COnline publication date: 1-May-2022
https://dl.acm.org/doi/10.1016/j.sysarc.2022.102476
Show More Cited By

Index Terms

TAFFO: Tuning Assistant for Floating to Fixed Point Optimization

Index terms have been assigned to the content through auto-classification.

Recommendations

Floating-point precision tuning using blame analysis
ICSE '16: Proceedings of the 38th International Conference on Software Engineering

While tremendously useful, automated techniques for tuning the precision of floating-point programs face important scalability challenges. We present Blame Analysis, a novel dynamic approach that speeds up precision tuning. Blame Analysis performs ...
Mizoguchi-Takahashi's type fixed point theorem

Recently, Eldred [A.A. Eldred, J. Anuradha, P. Veeramani, On equivalence of generalized multi-valued contactions and Nadler's fixed point theorem, J. Math. Anal. Appl. (2007) doi:10.1016/j.jmaa.2007.01.087] claimed that Nadler's [S.B. Nadler Jr., ...
Converses to fixed point theorems of Zermelo and Caristi

Let X be an abstract nonempty set and T be a self-map of X. Let Per T and Fix T denote the sets of all periodic points and all fixed points of T, respectively. Our main theorem says that if PerT = FixT≠θ, then there exists a partial ordering ≥ such that ...

Comments

Information & Contributors

Information

Published In

cover image IEEE Embedded Systems Letters

IEEE Embedded Systems Letters Volume 12, Issue 1

March 2020

32 pages

ISSN:1943-0663

Issue’s Table of Contents

1943-0663 © 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

Publisher

IEEE Press

Publication History

Published: 01 March 2020

Qualifiers

Research-article

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

10
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 26 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Xu JSong GZhou BLi FHao JZhao JLee IChabbi MSteuwer M(2024)A Holistic Approach to Automatic Mixed-Precision Code Generation and Tuning for Affine ProgramsProceedings of the 29th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming10.1145/3627535.3638484(55-67)Online publication date: 2-Mar-2024
https://dl.acm.org/doi/10.1145/3627535.3638484
Cattaneo DMaggioli AMagnani GDenisov LYang SAgosta GCherubin S(2023)Mixed Precision in Heterogeneous Parallel Computing Platforms via Delayed Code AnalysisEmbedded Computer Systems: Architectures, Modeling, and Simulation10.1007/978-3-031-46077-7_33(469-477)Online publication date: 2-Jul-2023
https://dl.acm.org/doi/10.1007/978-3-031-46077-7_33
Zoni DGalimberti A(2022)Cost-effective fixed-point hardware support for RISC-V embedded systemsJournal of Systems Architecture: the EUROMICRO Journal10.1016/j.sysarc.2022.102476126:COnline publication date: 1-May-2022
https://dl.acm.org/doi/10.1016/j.sysarc.2022.102476
Ferro QGraillat SHilaire TJézéquel FLewandowski B(2022)Neural Network Precision Tuning Using Stochastic ArithmeticSoftware Verification and Formal Methods for ML-Enabled Autonomous Systems10.1007/978-3-031-21222-2_10(164-186)Online publication date: 11-Aug-2022
https://dl.acm.org/doi/10.1007/978-3-031-21222-2_10
Aldinucci MAgosta GAndreini AArdagna CBartolini ACilardo ACosenza BDanelutto MEsposito RFornaciari WGiorgi RLengani DMontella ROlivieri MSaponara SSimoni DTorquati MPalesi MTumeo AGoumas GAlmudever C(2021)The Italian research on HPC key technologies across EuroHPCProceedings of the 18th ACM International Conference on Computing Frontiers10.1145/3457388.3458508(178-184)Online publication date: 11-May-2021
https://dl.acm.org/doi/10.1145/3457388.3458508
Cattaneo DChiari MFossati NCherubin SAgosta G(2021)Architecture-aware Precision Tuning with Multiple Number Representation Systems2021 58th ACM/IEEE Design Automation Conference (DAC)10.1109/DAC18074.2021.9586303(673-678)Online publication date: 5-Dec-2021
https://dl.acm.org/doi/10.1109/DAC18074.2021.9586303
Maier DJuurlink B(2021)Model-Based Loop PerforationEuro-Par 2021: Parallel Processing Workshops10.1007/978-3-031-06156-1_48(549-554)Online publication date: 30-Aug-2021
https://dl.acm.org/doi/10.1007/978-3-031-06156-1_48
Cherubin SCattaneo DChiari MAgosta G(2020)Dynamic Precision Autotuning with TAFFOACM Transactions on Architecture and Code Optimization10.1145/338878517:2(1-26)Online publication date: 29-May-2020
https://dl.acm.org/doi/10.1145/3388785
Cherubin SAgosta G(2020)Tools for Reduced Precision ComputationACM Computing Surveys10.1145/338103953:2(1-35)Online publication date: 17-Apr-2020
https://dl.acm.org/doi/10.1145/3381039
Baranowski MHe SLechner MNguyen TRakamarić Z(2020)An SMT Theory of Fixed-Point ArithmeticAutomated Reasoning10.1007/978-3-030-51074-9_2(13-31)Online publication date: 1-Jul-2020
https://dl.acm.org/doi/10.1007/978-3-030-51074-9_2

View Options

View options

Figures

Tables

Media

View Issue’s Table of Contents