Tau

I/O performance is becoming a key bottleneck in many cases at the extreme scale. As the volume of data and application reads and writes increases, it is important to assess the scalability of I/O operations as a key contributor to overall application performance. Optimizing I/O performance presents unique challenges for application developers and performance measurement and analysis tools, as it often involves an integration of multiple techniques to observe performance and create integrated I/O performance views from multiple I/O layers. Furthermore, reasoning about how to improve I/O performance necessarily involves understanding its relationship to the application behavior as a whole. The TAU Performance System [4, 7] is an integrated performance profiling and tracing toolkit that provides flexible support for performance instrumentation of both the application and the I/O layers.