Amazon cover image
Image from Amazon.com

Cooperative bug isolation : winning thesis of the 2005 ACM Doctoral Dissertation Competition / Ben Liblit.

By: Contributor(s): Material type: TextTextSeries: Lecture notes in computer science ; 4440. | ACM distinguished thesesPublication details: Berlin ; New York : Springer, ©2007.Description: 1 online resource (xv, 101 pages) : illustrationsContent type:
  • text
Media type:
  • computer
Carrier type:
  • online resource
ISBN:
  • 9783540718789
  • 3540718788
  • 354071877X
  • 9783540718772
  • 1280865725
  • 9781280865725
  • 9786610865727
  • 6610865728
Subject(s): Genre/Form: Additional physical formats: Print version:: Cooperative bug isolation.DDC classification:
  • 005 22
LOC classification:
  • QA76.6 .L5125 2007eb
Other classification:
  • TP31
Online resources:
Contents:
Instrumentation Framework -- Practical Considerations -- Techniques for Statistical Debugging -- Related Work -- Conclusion.
Summary: This monograph constitutes a thoroughly revised and extended version of the author's PhD thesis, which was selected as the winning thesis of the 2005 ACM Doctoral Dissertation Competition. Ben Liblit did his PhD work at the University of California, Berkeley, with Alexander Aiken as thesis adviser. This monograph reconsiders two common assumptions about how we should analyze software and arrives at some striking new results. This new approach makes use of some of the tools that biologists and economists use to understand their complicated systems by considering programs as statistical processes and using statistical techniques to understand software. The centerpiece of the monograph is an algorithm for isolating multiple bugs from sparsely sampled data taken from many thousands of program executions. This algorithm has unique properties that complement other program analysis techniques; in particular, it is potentially able to find the root cause of any program failure without first requiring an explicit specification of the property to check. The results Ben Liblit presents with his thesis represent a new and fundamental approach to software analysis and will provide a source of ideas and inspiration to the field for many years to come.
Holdings
Item type Current library Collection Call number Status Date due Barcode Item holds
eBook eBook e-Library eBook LNCS Available
Total holds: 0

Includes bibliographical references (pages 97-101).

Print version record.

Instrumentation Framework -- Practical Considerations -- Techniques for Statistical Debugging -- Related Work -- Conclusion.

This monograph constitutes a thoroughly revised and extended version of the author's PhD thesis, which was selected as the winning thesis of the 2005 ACM Doctoral Dissertation Competition. Ben Liblit did his PhD work at the University of California, Berkeley, with Alexander Aiken as thesis adviser. This monograph reconsiders two common assumptions about how we should analyze software and arrives at some striking new results. This new approach makes use of some of the tools that biologists and economists use to understand their complicated systems by considering programs as statistical processes and using statistical techniques to understand software. The centerpiece of the monograph is an algorithm for isolating multiple bugs from sparsely sampled data taken from many thousands of program executions. This algorithm has unique properties that complement other program analysis techniques; in particular, it is potentially able to find the root cause of any program failure without first requiring an explicit specification of the property to check. The results Ben Liblit presents with his thesis represent a new and fundamental approach to software analysis and will provide a source of ideas and inspiration to the field for many years to come.

English.

University staff and students only. Requires University Computer Account login off-campus.

Powered by Koha