This is an intermediate and advanced debugging tutorial focusing on ways to solve memory bugs in parallel and distributed applications. Memory problems such as heap memory leaks and array bounds violations can be challenging to track down in serial environments, and the challenges can be even more vexing when involving an MPI program running in a cluster environment. Tutorial participants will learn how to use Etnus TotalView to debug memory problems in both serial and parallel applications. In the process, they will learn to take advantage of heap error reporting, pointer annotation, and various report views through an interactive debugging session designed to track down memory problems. The tutorial with also focus on the use of heap status information -- to track down specific processes within large MPI applications for further scrutiny, and heap comparison features -- to make quick comparisons.

As developers address the shift towards multi-core processors they are encountering the challenges of debugging hybrid applications that are both multi-threaded and multi-process. Hybrid applications, which share memory among threads on a node but pass messages over the network from node to node, face the challenges of both deadlocks and race conditions. This talk will describe how to tackle these kinds of problems with the precise thread level control afforded by TotalView. Scientists and developers can use these tools and techniques to develop crucial insight into their application and quickly identify and resolve bugs.

TotalView is a flexible, scriptable parallel debugger with wide acceptance in the High Performance Computing community. This BOF will be an opportunity for TotalView users to share clever and interesting ways of adapting TotalView to their unique environment, using TotalView to do something unusual, or simply making the day to day process of debugging easier. Contact [email protected] if you want us to reserve time for you to tell your story or simply show up at the BOF and step forward.