The OpenTS parallel programming system

OpenTS is a parallel programming system for a wide spectrum of HPC hardware: multicore processors, SMPs, clusters and grids. OpenTS is the original Russian development that started in the Program Systems Institute of Russian Academy of Sciences (PSI RAS) more than 20 years ago. Currently OpenTS is being developed within the framework of supercomputing projects "SKIF" and "SKIF-GRID" of the Union State of Russia and Belarus.

OpenTS (T-system with open architecture) is the contemporary implementation of T-system concepts. It is noted for automatic dynamic parallelizing of programs. It provides a runtime for T++ high-level programming language that is a parallel dialect of C++.

OpenTS supports millions of "lightweight" threads of execution for each processor. Runtime environment provides very fast context switching between these threads that takes only a few nanoseconds. On top of this were implemented a lot of important elements and features: mobile threads, mobile objects and mobile references, distributed garbage collection, whiteboard that helps to carry out data and task exchanges, and so forth.

OpenTS has been successfully tested on a variety of scientific problems and on the HPC hardware of various scale: from multiprocessor computers to supercomputing installations of various architecture and productiveness.

If you have any questions, please contact us by e-mail supportATopentsDOTnet

Latest changes

2015/12/14: Released new OpenTS GNU/Linux packages.

2015/12/14: OpenTS v.1.4 released for Windows users.

2011/06/01: OpenTS v.1.3 released for Windows users.

2010/06/03: A new version of the OpenTS_VM released.

2010/03/03: OpenTS v.1.2 released for Windows users.