[Brown CS Talks] Brown CS Colloquium: Andreas Zeller in Room 368 (CIT 3rd floor) on 3/22/04 at 10:15 am

dvanhorn <[email protected]>
Newsgroups gmane.org.ballistichelmet.lambda
Message-ID <[email protected]>

-------- Original Message --------
Subject: [Brown CS Talks] Brown CS Colloquium:  Andreas Zeller in Room 368 
(CIT 3rd floor) on 3/22/04 at 10:15 am
Date: Wed, 3 Mar 2004 13:37:03 -0500
From: [email protected]
Reply-To: [email protected]
To: <[email protected]>

			   BROWN UNIVERSITY

		     COMPUTER SCIENCE COLLOQUIUM


			    Andreas Zeller

			 Saarland University


		  Monday, March 22, 2004 at 10:15 am

		       Room 368 (CIT 3rd floor)


		      Why Does My Program Fail?


Despite all advances in program analysis, narrowing down failure
causes is still a manual task, dependent on the skill and the
endurance of the programmer.

In this talk, I present some techniques that widely automate the quest
for failure causes.  Based on a simple trial-and-error process and an
automated test, it is fairly simple to isolate external failure causes
like input, code differences, or thread schedules.  Advanced
techniques can even isolate cause-effect chains within the program
run: ``First, variable x_1 had a value of x_1, thus v_2 became x_2,
thus v_3 was set to x_3 ... and thus the program failed''.  Case
studies on real programs with real bugs, such as the GNU C compiler,
demonstrate the practical feasibility of these experimental
approaches.

Already today, anyone can submit buggy programs via our ``AskIgor''
web service to obtain failure diagnoses automatically; integration
into common programming environments is underway.  In the future, user
applications may leverage automated diagnosis to become
self-diagnosing and even partially self-healing - just by linking in a
special diagnosis library.

Andreas Zeller is a computer science professor at Saarland University,
Saarbruecken, Germany, since 2001.  His research concerns the analysis
of large software systems and their histories, especially the analysis
of why these systems fail to work as they should.

  			
		Host: Professor Shriram Krishnamurthi


_______________________________________________
The Computer Science Department is located at 115 Waterman St., Providence, 
RI.  Our Calendar of Events at http://www.cs.brown.edu/events lists all of our 
talks.

If you would like to be removed from (or added to) our seminar mailing list, 
or if you would prefer to receive only announcements pertaining to IPP talks, 
please send email containing your request to mailto:[email protected]

We'd very much appreciate your forwarding these announcements to colleagues 
who may also have an interest.  Thanks for your cooperation.
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.