Schwarz Library

Schwarz Library is an implementation of the asynchronous Schwarz methods using Ginkgo.

View the Project on GitHub pratikvn/schwarz-lib

Schwarz Library

Build status Documentation

Performance results

  1. Paper in IJHPCA; Alternative arXiv version

Required components

The required components include:

  1. Ginkgo: The Ginkgo library is needed. It needs to be installed and preferably the installation path provided as an environment variable in Ginkgo_DIR variable.
  2. MPI: As multiple nodes and a domain decomposition is used, an MPI implementation is necessary.

Quick Install

Building Schwarz-Lib

To build Schwarz-Lib, you can use the standard CMake procedure.

mkdir build; cd build
cmake -G "Unix Makefiles" .. && make

By default, SCHWARZ_BUILD_BENCHMARKING is enabled. This allows you to quickly run an example with the timings if needed. For a detailed list of options available see the Benchmarking page.

For more CMake options please refer to the Installation page

Currently implemented features

  1. Executor paradigm:
    • GPU.
    • OpenMP.
    • Single rank per node and threading in one node.
  2. Factorization paradigm:
    • CHOLMOD.
    • UMFPACK.
  3. Solving paradigm:
    • Direct:
    • Ginkgo.
    • CHOLMOD.
    • UMFPACK.
    • Iterative:
    • Ginkgo.
    • deal.ii.
  4. Partitioning paradigm:
    • METIS.
    • Regular, 1D.
    • Regular, 2D.
    • Zoltan.
  5. Convergence check:
    • Centralized, tree based convergence (Yamazaki 2019).
    • Decentralized, leader election based (Bahi 2005).
  6. Communication paradigm.
    • Onesided.
    • Twosided.
    • Event based.
  7. Communication strategies.
    • Remote comm strategies:
    • MPI_Put , gathered.
    • MPI_Put , one by one.
    • MPI_Get , gathered .
    • MPI_Get , one by one. + [x] Lock strategies: MPI_Win_lock / MPI_Win_lock_all .
    • Lock all and unlock all.
    • Lock local and unlock local. + [x] Flush strategies: MPI_Win_flush / MPI_Win_flush_local .
    • Flush all.
    • Flush local.
  8. Schwarz problem type.
    • RAS.
    • O-RAS.

Any of the implemented features can be permuted and tested.

Known Issues

  1. On Summit, the Spectrum MPI seems to have a bug with using MPI_Put with GPU buffers. MPI_Get works as expected. This bug has also been confirmed with an external micro-benchmarking library, OSU Micro-Benchmarks.

For installing and building, please check the Installation page

Credits: This code was inspired by (written in C++, with additions and improvements) the code from Ichitaro Yamazaki, ICL, UTK.