Citation
Hiren Patel. "Predictable Programming on a Precision Timed Architecture". Talk or presentation, 21, October, 2008.

Abstract
In a hard real-time embedded system the time at which a result is computed is as important as the result itself. Modern processors go to extreme lengths to ensure their function is predictable but have abandoned predictable timing in favor of average-case performance. Real-time operating systems provide timing-aware scheduling policies but without precise worst-case execution time bounds they cannot provide guarantees. We describe an alternative: a SPARC-based processor with predictable timing and instruction-set extensions that provide precise timing control. Its pipeline executes multiple independent hardware threads to avoid costly unpredictable bypassing and its exposed memory hierarchy provides predictable latency. We demonstrate the effectiveness of this precision-timed (PRET) architecture through example applications running in simulation.

Electronic downloads

• cases08-pret-final.ppt · application/vnd.ms-powerpoint · 2540 kbytes

• HTML

  Hiren Patel. <a
  href="http://chess.eecs.berkeley.edu/pubs/518.html"
  ><i>Predictable Programming on a Precision Timed
  Architecture</i></a>, Talk or presentation,  21,
  October, 2008.

• Plain text

  Hiren Patel. "Predictable Programming on a Precision
  Timed Architecture". Talk or presentation,  21,
  October, 2008.

• BibTeX

  @presentation{Patel08_PredictableProgrammingOnPrecisionTimedArchitecture,
      author = {Hiren Patel},
      title = {Predictable Programming on a Precision Timed
                Architecture},
      day = {21},
      month = {October},
      year = {2008},
      abstract = {In a hard real-time em<x>bedded system the time at
                which a result is computed is as important as the
                result itself. Modern processors go to extreme
                lengths to ensure their function is predictable
                but have abandoned predictable timing in favor of
                average-case performance. Real-time operating
                systems provide timing-aware scheduling policies
                but without precise worst-case execution time
                bounds they cannot provide guarantees. We describe
                an alternative: a SPARC-ba<x>sed processor with
                predictable timing and instruction-set extensions
                that provide precise timing control. Its pipeline
                executes multiple independent hardware threads to
                avoid costly unpredictable bypassing and its
                exposed memory hierarchy provides predictable
                latency. We demonstrate the effectiveness of this
                precision-timed (PRET) architecture through
                example applications running in simulation.},
      URL = {http://chess.eecs.berkeley.edu/pubs/518.html}
  }
  

Posted by Hiren Patel on 21 Jan 2009.
Groups: pret
For additional information, see the Publications FAQ or contact webmaster at chess eecs berkeley edu.