Version 6.0 of SCOTCH was released 2 December 2012. 31 December 2021 -- Version 7.0 of SCOTCH is out. Update it, please.
(Ответ для Iakunin Andrei на комментарий #0) > Version 6.0 of SCOTCH was released 2 December 2012. > > 31 December 2021 -- Version 7.0 of SCOTCH is out. > > Update it, please. Why do you need it?
Useful for depended software. Many updates. This package used in OpenFOAM, one of the biggest open source CFD program. Is this a reason to update from 12 years release? Main changes since 5.x version: Version 7.0 of SCOTCH release (codename « Sankara »), fruition of six years of development, brings many major improvements. Notably: 1) Thread management is now fully dynamic: Scotch automatically adapts its behavior to the number of available cores. An existing thread pool can be imported from a caller software, so as not to create unnecessary threads and to improve computation locality. 2) Many algorithms have been rewritten to benefit from shared-memory parallelism, including some algorithms of PT-Scotch, which now benefit from a hybrid « MPI+threads » parallelism. 3) Execution options, such as the fact to select between deterministic algorithms that allow for computational reproducibility or faster, non-deterministic algorithms, can now be selected dynamically and no longer at compile time. 4) A new class of objects is available: execution contexts. Contexts allow users to encapsulate a thread pool, execution options, and a private pseudo-random generator, so as to execute concurrently (and in a fully reproducible way) any Scotch computation (mapping, partitioning, ordering, etc.) on the same graph or on different graphs. Version 6.0 of SCOTCH offers many new features: sequential graph repartitioning; sequential graph partitioning with fixed vertices; sequential graph repartitioning with fixed vertices; new, fast, direct k-way partitioning and mapping algorithms; multi-threaded, shared memory algorithms in the (formerly) sequential part of the library; exposure in the API of many distributed graph handling routines; embedded pseudo-random generator for improved reproducibility; and even more...
