Thursday, October 23, 2008
12:00 pm - 1:00 pm
PLACE: Wean Hall 8220
Precise Causal Traces of Black-box Applications with BorderPatrol
Causal request traces are valuable to developers of large concurrent and distributed applications, yet difficult to obtain. Traces show how a request is processed, and can be analyzed by tools to detect performance or correctness errors and anomalous behavior.
BorderPatrol obtains precise request traces through systems built from a litany of unmodified modules. Traced components include Apache, thttpd, PostgreSQL, TurboGears, BIND and notably Zeus, a closed-source event-driven web server. BorderPatrol obtains traces using active observation which carefully modifies the event stream observed by modules, simplifying precise observation. Protocol processors leverage knowledge about standard protocols, avoiding application-specific instrumentation. Using these techniques BorderPatrol obtains precise traces for black-box systems that cannot be traced by any other technique.
John Jannotti is an Assistant Professor of Computer Science at Brown University. At Brown, John's research includes simplifying the design of large-scale distributed systems, accountable yet anonymous peer-to-peer systems, and smart camera networks. Before joining Brown, John received his PhD from the Massachusetts Institute of Technology. While at MIT, John worked on operating system architectures, software routers, wireless protocols, and next-generation networking designs.
Visitor Host: David Anderson
Visitor Coordinator: Angie Miller
Karen Lindenfelser (8-6716)
or visit http://www.pdl.cmu.edu/SDI/