| IBM researchers | University researchers |
| Satish Chandra Stephen Fink Edith Schonberg Gary Sevitsky Frank Tip | Rastislav Bodik, UC Berkeley Bruno Dufour, Rutgers University Barbara Ryder, Rutgers University Manu Sridharan, UC Berkeley |
Learn more about the IBM-Rutgers-Berkeley collaboration in these podcast presentations. Transcripts available.
- How can we find the sources of bloat in applications? IBM's Gary Sevitsky talks about fixing an application's problematic "hot areas." 8 min., 32 sec.
- How overuse of temporary objects hampers software performance. Barbara Ryder, Rutgers University, talks about working on a blended analysis project with IBM to improve software quality. 9 min., 49 sec.
- How open source helps developers debug software. IBM's Stephen Fink talks about his work with computer scientists at the University of California at Berkeley to analyze what gets in the way of superior software functionality. 5 min., 18 sec.
Results
- Ras Bodik is getting ready to release open-source software under the Eclipse public license to the T.J. Watson Libraries for Analysis on Sourceforge.
- In July 2007, Rutgers' Barbara Ryder and IBM's Frank Tip released an executable version of Chianti, a change impact analysis tool that lets programmers examine the effects of edits they make to Java code. If a test fails after an edit, Chianti highlights the code changes that may have affected that test and facilitates exploration of these changes to find which ones induced the failure.
- Crisp - A fault localization tool for Java programs (abstract and demo). Ophelia Chesley; Xiaoxia Ren, Barbara G. Ryder and Frank Tip. Proceedings of the 29th International Conference on Software Engineering, May 2007.
- Blended analysis for performance understanding of framework-based applications (pdf), ISSTA’07, London. July 9–12, 2007. Barbara Ryder, Gary Sevitsky and Bruno Dufour define a new analysis paradigm -- blended program analysis -- that enables practical, effective analysis of large framework-based Java applications for performance understanding.
- Thin Slicing (pdf), PLDI 2007, UC, San Diego. June 11-13, 2007. IBM's Stephen Fink and UC Berkeley's Rastislav Bodik and Manu Sridharan demonstrate a simple but effective dependence analysis for software debugging and program-understanding tasks.
For details about OCR, contact Steve Lavenberg: sslaven@us.ibm.com
Last updated on December 7, 2007