skip to main content
10.1145/1837274.1837492acmconferencesArticle/Chapter ViewAbstractPublication PagesdacConference Proceedingsconference-collections
research-article

Best-effort computing: re-thinking parallel software and hardware

Published: 13 June 2010 Publication History

Abstract

With the advent of mainstream parallel computing, applications can obtain better performance only by scaling to platforms with larger numbers of cores. This is widely considered to be a very challenging problem due to the difficulty of parallel programming and the bottlenecks to efficient parallel execution. Inspired by how networking and storage systems have scaled to handle very large volumes of packet traffic and persistent data, we propose a new approach to the design of scalable, parallel computing platforms. For decades, computing platforms have gone to great lengths to ensure that every computation specified by applications is faithfully executed. While this design philosophy has remained largely unchanged, applications and the basic characteristics of their workloads have changed considerably. A wide range of existing and emerging computing workloads have an inherent forgiving nature. We therefore argue that adopting a best-effort service model for various software and hardware components of the computing platform stack can lead to drastic improvements in scalability. Applications are cognizant of the best-effort model, and separate their computations into those that may be executed on a best-effort basis and those that require the traditional execution guarantees. Best-effort computations may be exploited to simply reduce the computing workload, shape it to be more suitable for parallel execution, or execute it on unreliable hardware components. Guaranteed computations are realized either through an overlay software layer on top of the best-effort substrate, or through the use of application-specific strategies. We describe a system architecture for a best-effort computing platform, provide examples of parallel software and hardware that embody the best-effort model, and show that large improvements in performance and energy efficiency are possible through the adoption of this approach.

References

[1]
S. Floyd and M. Allman. Comments on the Usefulness of Simple Best-Effort Traffic. IETF RFC 5290, http://tools.ietf.org/html/rfc5290, July 2008.
[2]
B. Fitzpatrick. Livejournal: Behind the scenes scaling storytime (Invited talk). In USENIX, 2007.
[3]
G. Linden. Early Amazon: Splitting the website. Online: http://glinden.blogspot.com/2006/02/early-amazon-splitting-web.
[4]
J. Newton. Scaling out like technorati. http://newton.typepad.com/content/2007/09/scaling-outlik.html.
[5]
T. O'Reilly. Web 2.0 and databases part 1: Second life. http://radar.oreilly.com/archives/2006/04/web-20-and-databases-part-1-se.html.
[6]
T. O'Reilly. Database war stories 5: craigslist. http://radar.oreilly.com/archives/2006/04/database-war-stories-5-craigsl.html.
[7]
T. O'Reilly. Database war stories 3: Flickr. http://radar.oreilly.com/archives/2006/04/database-war-stories-3-flickr.html.
[8]
M. Armbrust et al. SCADS: Scale-independent storage for social computing applications. In Proc. 4th Biennial Conference on Innovative Data Systems Research (CIDR), 2009.
[9]
Adf Facebook infosession. Sponsored by Industrial Relations Office (IRO), EECS Department, UC Berkeley, September 2008.
[10]
J. Sobel. Scaling out. http://www.facebook.com/note.php?note_id=23844338919.
[11]
S. Armstrong et al. Multicast Transport Protocol. RFC 1301, http://www.rfc-archive.org/getrfc.php?rfc=1301, Feb. 1992.
[12]
L. Breslau and S. Shenker. Best-effort versus reservations: a simple comparative analysis. In Proc. SIGCOMM, pages 3--16, 1998.
[13]
Pradeep Dubey. A Platform 2015 Workload Model: Recognition, Mining and Synthesis Moves Computers to the Era of Tera. White Paper, Intel Corporation, 96, 2008.
[14]
Y. K. Chen et al. Convergence of Recognition, Mining, and Synthesis Workloads and Its Implications. Proc. of IEEE, 96:790--807, 2008.
[15]
J. B. MacQueen. Some methods for classification and analysis of multivariate observation. In Proceedings of the Berkeley Symposium on Mathematical Statistics and Probability, 1967.
[16]
J. Meng, S. Chakradhar, and A. Raghunathan. Best-effort parallel execution framework for recognition and mining applications. In Proc. IEEE Int. Parallel and Distributed Processing Symp., May 2009.
[17]
J. Meng, A. Raghunathan, A. Chakradhar, and S. Byna. Exploiting the forgiving nature of applications for scalable parallel execution. In Proc. IEEE Int. Parallel and Distributed Processing Symp., Apr. 2010.
[18]
S. Byna, J. Meng, A. Raghunathan, S. Chakradhar, and S. Cadambi. Best Effort Semantic Document Search on GPUs. In Proc. Third Wkshp. on General-Purpose Computation on Graphics Procesing Units, Mar. 2010.
[19]
V. Chippa, D. Mohapatra, A. Raghunathan, K. Roy, and S. T. Chakradhar. Scalable Effort Hardware: Exploiting Algorithmic Resilience for Energy Efficiency. In Prof. ACM/IEEE Design Automation Conf., June 2010.
[20]
N. R. Pal, J. C. Bezdek, and E. C.-K. Tsao. Generalized clustering networks and Kohonen's self-organizing scheme. IEEE Trans. on Neural Networks, 4(4):549--557, Jul 1993.
[21]
L. Leem, H. Cho, J. Bau, Q. Jacobson, and S. Mitra. ERSA: Error-Resilient System Architecture for Probabilistic Applications. In Prof. ACM/IEEE Design Automation and Test in Europe, Mar. 2010.
[22]
H. Cho, S. T. Chakradhar, and A. Raghunathan. Trading reliability for parallelism. Personal Communication, Sep. 2009.

Cited By

View all
  • (2025)Designing Approximate Reduced Complexity Wallace MultipliersElectronics10.3390/electronics1402033314:2(333)Online publication date: 16-Jan-2025
  • (2023)A feasibility analysis of image approximation with image quality assessmentsIET Image Processing10.1049/ipr2.12993Online publication date: 14-Nov-2023
  • (2023)Performance Efficient and Fault Tolerant Approximate AdderJournal of Electronic Testing10.1007/s10836-023-06092-539:5-6(571-582)Online publication date: 11-Dec-2023
  • Show More Cited By

Index Terms

  1. Best-effort computing: re-thinking parallel software and hardware

    Recommendations

    Comments

    Information & Contributors

    Information

    Published In

    cover image ACM Conferences
    DAC '10: Proceedings of the 47th Design Automation Conference
    June 2010
    1036 pages
    ISBN:9781450300025
    DOI:10.1145/1837274
    Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]

    Sponsors

    Publisher

    Association for Computing Machinery

    New York, NY, United States

    Publication History

    Published: 13 June 2010

    Permissions

    Request permissions for this article.

    Check for updates

    Author Tags

    1. best effort systems
    2. multi core
    3. parallel computing
    4. performance
    5. scalability

    Qualifiers

    • Research-article

    Conference

    DAC '10
    Sponsor:

    Acceptance Rates

    Overall Acceptance Rate 1,770 of 5,499 submissions, 32%

    Upcoming Conference

    DAC '25
    62nd ACM/IEEE Design Automation Conference
    June 22 - 26, 2025
    San Francisco , CA , USA

    Contributors

    Other Metrics

    Bibliometrics & Citations

    Bibliometrics

    Article Metrics

    • Downloads (Last 12 months)11
    • Downloads (Last 6 weeks)0
    Reflects downloads up to 20 Jan 2025

    Other Metrics

    Citations

    Cited By

    View all
    • (2025)Designing Approximate Reduced Complexity Wallace MultipliersElectronics10.3390/electronics1402033314:2(333)Online publication date: 16-Jan-2025
    • (2023)A feasibility analysis of image approximation with image quality assessmentsIET Image Processing10.1049/ipr2.12993Online publication date: 14-Nov-2023
    • (2023)Performance Efficient and Fault Tolerant Approximate AdderJournal of Electronic Testing10.1007/s10836-023-06092-539:5-6(571-582)Online publication date: 11-Dec-2023
    • (2022)As-Is Approximate ComputingACM Transactions on Architecture and Code Optimization10.1145/355976120:1(1-26)Online publication date: 17-Nov-2022
    • (2022)Probability-Driven Evaluation of Lower-Part Approximation AddersIEEE Transactions on Circuits and Systems II: Express Briefs10.1109/TCSII.2021.309398469:1(204-208)Online publication date: Jan-2022
    • (2022)Reduced Precision DWC: An Efficient Hardening Strategy for Mixed-Precision ArchitecturesIEEE Transactions on Computers10.1109/TC.2021.305887271:3(573-586)Online publication date: 1-Mar-2022
    • (2022)LLVM-ACT: Profiling Based Tool for Approximate Computing Technique Selection2022 XII Brazilian Symposium on Computing Systems Engineering (SBESC)10.1109/SBESC56799.2022.9965085(1-8)Online publication date: 21-Nov-2022
    • (2022)A Survey on Machine Learning Accelerators and Evolutionary Hardware PlatformsIEEE Design & Test10.1109/MDAT.2022.316112639:3(91-116)Online publication date: Jun-2022
    • (2022)Approximate Computing at the Algorithmic LevelApproximate Computing Techniques10.1007/978-3-030-94705-7_5(109-142)Online publication date: 3-Jan-2022
    • (2022)Approximate function memoizationConcurrency and Computation: Practice and Experience10.1002/cpe.720434:23Online publication date: 20-Jul-2022
    • Show More Cited By

    View Options

    Login options

    View options

    PDF

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader

    Media

    Figures

    Other

    Tables

    Share

    Share

    Share this Publication link

    Share on social media