Rogue Wave Software has announced that TotalView has been selected by CSC – IT Center for Science to debug scientific and research applications on its new Cray XC30 supercomputer, Sisu.
Offering IT support for academia, research institutes, and companies, CSC provides Finland's widest selection of scientific software and databases, as well as the country’s most powerful supercomputing environment, which is also available to researchers via the Funet network. CSC researchers and post-doctorate scientists will use TotalView to help develop and debug applications ranging from quantum physics and astrophysics, nano-sciences and chemistry to large-scale meteorological and climatological applications.
Many of these applications, which require reliable and flawless execution over many hours when run in parallel on hundreds of MPI-tasks, are part of CSC’s Grand Challenge programs. Aimed at high-impact scientific research, these programs require computational or data resources exceeding CSC's standard project quotas or level of services. The TotalView debugger will be deployed on CSC’s HP-clusters and the Cray XC30 supercomputer, which is the first Cray supercomputer to use Intel CPUs.
‘The CSC continues to leverage Rogue Wave’s tools because they are invaluable to our scientists and researchers. Our users are very impressed with TotalView’s easy-to-use and intuitive features and they are excited about the new reverse debugging capabilities available on the Cray, which will significantly decrease debugging time,’ commented Sami Saarinen, senior HPC applications specialist at CSC - IT Center for Science. ‘While we thoroughly investigated alternative debuggers, we found the competition to be far behind TotalView’s features.’
TotalView is a scalable and intuitive debugger for parallel applications written in C, C++ and Fortran. Designed to improve developer productivity, TotalView simplifies and shortens the process of developing, debugging and optimising complex applications. It provides a powerful combination of capabilities for pinpointing and fixing hard-to-find bugs, such as race conditions, memory leaks, and memory overruns.