*banner
 

Predictability, Repeatability, and Models for Cyber-Physical Systems
Edward A. Lee

Citation
Edward A. Lee. "Predictability, Repeatability, and Models for Cyber-Physical Systems". Talk or presentation, 24, October, 2010; Invited talk, Workshop on Foundations of Component Based Design (WFCD) at ESWeek 2010, Scottsdale, AZ.

Abstract
To be able to analyze and control the joint dynamics of software, networks, and physical processes, engineers need to be able to analyze and control the temporal behavior of software and networks. Most real-time systems techniques aim to do this by focusing on predicting execution time bounds through program analysis and architectural modeling. In this talk, I attempt to give a rigorous meaning to "prediction" in this context, and argue that what engineers need is not predictability so much as repeatability. The argument relies on the observation that Boolean properties (properties that are either satisfied or not satisfied) are necessarily properties of models and not properties of the physical realizations of systems. Predictability is about the ability to anticipate whether properties will be satisfied, whereas repeatability is about the ability of a physical realization to deliver consistent behavior.

Electronic downloads

Citation formats  
  • HTML
    Edward A. Lee. <a
    href="http://chess.eecs.berkeley.edu/pubs/798.html"
    ><i>Predictability, Repeatability, and Models for
    Cyber-Physical Systems</i></a>, Talk or
    presentation,  24, October, 2010; Invited talk, Workshop on
    Foundations of Component Based Design (WFCD) at ESWeek 2010,
    Scottsdale, AZ.
  • Plain text
    Edward A. Lee. "Predictability, Repeatability, and
    Models for Cyber-Physical Systems". Talk or
    presentation,  24, October, 2010; Invited talk, Workshop on
    Foundations of Component Based Design (WFCD) at ESWeek 2010,
    Scottsdale, AZ.
  • BibTeX
    @presentation{Lee10_PredictabilityRepeatabilityModelsForCyberPhysical,
        author = {Edward A. Lee},
        title = {Predictability, Repeatability, and Models for
                  Cyber-Physical Systems},
        day = {24},
        month = {October},
        year = {2010},
        note = {Invited talk, Workshop on Foundations of Component
                  Based Design (WFCD) at ESWeek 2010, Scottsdale, AZ},
        abstract = {To be able to analyze and control the joint
                  dynamics of software, networks, and physical
                  processes, engineers need to be able to analyze
                  and control the temporal behavior of software and
                  networks. Most real-time systems techniques aim to
                  do this by focusing on predicting execution time
                  bounds through program analysis and architectural
                  modeling. In this talk, I attempt to give a
                  rigorous meaning to "prediction" in this context,
                  and argue that what engineers need is not
                  predictability so much as repeatability. The
                  argument relies on the observation that Boolean
                  properties (properties that are either satisfied
                  or not satisfied) are necessarily properties of
                  models and not properties of the physical
                  realizations of systems. Predictability is about
                  the ability to anticipate whether properties will
                  be satisfied, whereas repeatability is about the
                  ability of a physical realization to deliver
                  consistent behavior.},
        URL = {http://chess.eecs.berkeley.edu/pubs/798.html}
    }
    

Posted by Christopher Brooks on 22 Dec 2010.
For additional information, see the Publications FAQ or contact webmaster at chess eecs berkeley edu.

Notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright.

©2002-2018 Chess