|Livermore Computing Workshop Announcement|
|Title:||Performance Analysis and Tuning Parallel Codes with HPCToolkit|
|*Note: must be scheduled through Martin Schulz email@example.com|
|Description:||Rice University's HPCToolkit is an integrated suite of tools for measurement and analysis of program performance on computers ranging from multicore desktop systems to supercomputers. By using asynchronous sampling of timers and hardware performance counters, HPCToolkit collects accurate measurements of a program's work, resource consumption, and inefficiency; it attributes these metrics to full calling contexts. HPCToolkit works with multilingual, fully-optimized applications that are statically or dynamically linked. Since HPCToolkit uses sampling, it has low measurement overhead (typically < 5% for sampling frequencies of up to 1000/second) and scales to large parallel systems. HPCToolkit includes presentation tools enable rapid analysis of a program's execution costs, inefficiency, and scaling characteristics both within and across nodes of a parallel system. HPCToolkit supports measurement and analysis of serial codes, threaded codes (e.g. pthreads, OpenMP), MPI, and hybrid (MPI+threads) parallel codes. HPCToolkit's graphical user interfaces support code-centric, thread-centric and time-centric analysis. Emerging capabilities in HPCToolkit include support for parallel systems equipped with accelerators such as GPUs, and enhanced support for measurement and analysis of applications that employ OpenMP. For more information on HPCToolkit see hpctoolkit.org.|
|Presentors:||John Mellor-Crummey and Mark Krentel, Rice University|
|Registration & Questions:||The Aug 7, 10am-noon General Presentation session is open to all without registration.
The Hands-on Lab session and individual/code team sessions require registration and scheduling by contacting Martin Schulz (firstname.lastname@example.org).