Yılmaz, Cemal and Porter, Adam (2010) Combining hardware and software instrumentation to classify program executions. In: International Symposium on the Foundations of Software Engineering (FSE), (Accepted/In Press)
![[thumbnail of cyilmaz.pdf]](https://research.sabanciuniv.edu/style/images/fileicons/application_pdf.png)
cyilmaz.pdf
Download (259kB)
Abstract
Several research efforts have studied ways to infer properties of software systems from program spectra gathered from the running systems, usually with software-level instrumentation. While these efforts appear to produce accurate classifications, detailed understanding of their costs and potential cost-benefit tradeoffs is lacking. In this work we present a hybrid instrumentation approach which uses hardware performance counters to gather program spectra at very low cost. This underlying data is further augmented with data captured by minimal amounts of software-level instrumentation. We also
evaluate this hybrid approach by comparing it to other existing approaches. We conclude that these hybrid spectra can reliably distinguish failed executions from successful executions at a fraction of the runtime overhead cost of using software-based execution data.
Item Type: | Papers in Conference Proceedings |
---|---|
Subjects: | Q Science > Q Science (General) |
Divisions: | Faculty of Engineering and Natural Sciences > Academic programs > Computer Science & Eng. Faculty of Engineering and Natural Sciences |
Depositing User: | Cemal Yılmaz |
Date Deposited: | 18 May 2010 14:37 |
Last Modified: | 26 Apr 2022 08:55 |
URI: | https://research.sabanciuniv.edu/id/eprint/13962 |