Hiop Versions Save

What's Changed

• Update modules for CI tests on LLNL LC by @nychiang in https://github.com/LLNL/hiop/pull/679
• Update cmake build system to require RAJA when GPU compute mode is used by @nychiang in https://github.com/LLNL/hiop/pull/676
• Moving limits options for NLP IPM solvers by @cnpetra in https://github.com/LLNL/hiop/pull/681

Full Changelog: https://github.com/LLNL/hiop/compare/v1.0.2...v1.0.3

What's Changed

• Removed deprecated ALG2 for cusparseCsr2cscEx2 by @cnpetra in https://github.com/LLNL/hiop/pull/671
• Addressed fixed buffer size vulnerability for vsnprintf by @nychiang in https://github.com/LLNL/hiop/pull/673
• Removed stringent -Wall and -Werror from release builds to avoid downstream compilation errors

Full Changelog: https://github.com/LLNL/hiop/compare/v1.0.1...v1.0.2

What's Changed

Default C++ standard remains C++14

• Fix LLNL CI by @nychiang in https://github.com/LLNL/hiop/pull/663
• C++17 support: compile -Wall/-Werror proof by @tepperly in https://github.com/LLNL/hiop/pull/653
• Fix a bug in copying an empty matrix into a bigger matrix by @nychiang in https://github.com/LLNL/hiop/pull/666
• Fix the approach used to update mu by @nychiang in https://github.com/LLNL/hiop/pull/664

Full Changelog: https://github.com/LLNL/hiop/compare/v1.0.0...v1.0.1

Notable new features

Interfaces of various solvers reached an equilibrium point after HiOp was interfaced with multiple optimization front-ends (e.g., power grid ACOPF and SC-ACOPF problems and topology optimization) both on CPUs and GPUs. The PriDec solver reached exascale on Frontier after minor communication optimizations. The quasi-Newton interior-point solver received a couple of updates that increase robustness. The Newton interior-point solver can fully operate on GPUs with select GPU linear solvers (CUSOLVER-LU and Gingko).

• Instrumentation of RAJA sparse matrix class with execution spaces by @cnpetra in https://github.com/LLNL/hiop/pull/589
• Fix Assignment Typo in hiopMatrixSparseCsrCuda.cpp by @pate7 in https://github.com/LLNL/hiop/pull/612
• Use failure not failed in PNNL commit status posting by @cameronrutherford in https://github.com/LLNL/hiop/pull/609
• rebuild modules on quartz by @nychiang in https://github.com/LLNL/hiop/pull/619
• Use constraint violation in checkTermination by @nychiang in https://github.com/LLNL/hiop/pull/617
• MPI communication optimization by @rothpc in https://github.com/LLNL/hiop/pull/613
• fix memory leaks in inertia-free alg and condensed linsys by @nychiang in https://github.com/LLNL/hiop/pull/622
• Update IPM algorithm for the dense solver by @nychiang in https://github.com/LLNL/hiop/pull/616
• Use integer preprocessor macros for version information by @tepperly in https://github.com/LLNL/hiop/pull/627
• use compound vec in bicg IR by @nychiang in https://github.com/LLNL/hiop/pull/621
• Use bicg ir in the quasi-Newton solver by @nychiang in https://github.com/LLNL/hiop/pull/620
• Add support to MPI in C/Fortran examples by @nychiang in https://github.com/LLNL/hiop/pull/633
• Refactor CUSOLVER-LU module and interface by @pelesh in https://github.com/LLNL/hiop/pull/634
• Add MPI unit test for DenseEx4 by @nychiang in https://github.com/LLNL/hiop/pull/644
• Add more options to control NLP scaling by @nychiang in https://github.com/LLNL/hiop/pull/649
• Development of the feasibility restoration in the quasi-Newton solver by @nychiang in https://github.com/LLNL/hiop/pull/647
• GPU linear solver interface by @pelesh in https://github.com/LLNL/hiop/pull/650

New Contributors

• @pate7 made their first contribution in https://github.com/LLNL/hiop/pull/612
• @rothpc made their first contribution in https://github.com/LLNL/hiop/pull/613

This release hosts a series of comprehensive internal developments and software re-engineering to improve the portability and performance on accelerators/GPU platforms. No changes to the user interface permeated under this release.

Notable new features

A new execution space abstraction is introduced to allow multiple hardware backends to run concurrently. The proposed design differentiates between "memory backend" and "execution policies" to allow using RAJA with Umpire-managed memory, RAJA with Cuda- or Hip-managed memory, RAJA with std memory, Cuda/Hip kernels with Cuda-/Hip- or Umpire-managed memory, etc.

• Execution spaces: support for memory backends and execution policies by @cnpetra in https://github.com/LLNL/hiop/pull/543
• Build: Cuda without raja by @cnpetra in https://github.com/LLNL/hiop/pull/579
• Update of RAJA-based dense matrix to support runtime execution spaces by @cnpetra in https://github.com/LLNL/hiop/pull/580
• Reorganization of device namespace by @cnpetra in https://github.com/LLNL/hiop/pull/582
• RAJA Vector int with ExecSpace by @cnpetra in https://github.com/LLNL/hiop/pull/583
• Instrumentation of host vectors with execution spaces by @cnpetra in https://github.com/LLNL/hiop/pull/584
• Remove copy from/to device methods in vector classes by @cnpetra in https://github.com/LLNL/hiop/pull/587
• Add support for Raja with OpenMP into LLNL CI by @nychiang in https://github.com/LLNL/hiop/pull/566

New vector classes using vendor-provided API were introduced and documentation was updated/improved

• Development of hiopVectorCuda by @nychiang in https://github.com/LLNL/hiop/pull/572
• Implementation of hiopVectorHip by @nychiang in https://github.com/LLNL/hiop/pull/590
• Update user manual by @nychiang in https://github.com/LLNL/hiop/pull/591
• Update the code comments in hiopVector classes by @nychiang in https://github.com/LLNL/hiop/pull/592

Refinement of triangular solver implementation for Ginkgo by @fritzgoebel in https://github.com/LLNL/hiop/pull/585

Bug fixes

• Refine the computation in normal equation system by @nychiang in https://github.com/LLNL/hiop/pull/530
• Fix static culibos issue #567 by @nychiang in https://github.com/LLNL/hiop/pull/568
• Fix segfault, remove nonsymmetric ginkgo solver by @fritzgoebel in https://github.com/LLNL/hiop/pull/548
• Calculate the inverse objective scale correctly. by @tepperly in https://github.com/LLNL/hiop/pull/570
• Fix hiopVectorRajaPar::copyToStartingAt_w_pattern by @nychiang in https://github.com/LLNL/hiop/pull/569
• Gitlab pipeline refactor by @CameronRutherford in https://github.com/LLNL/hiop/pull/597

New Contributors

• @tepperly made their first contribution in https://github.com/LLNL/hiop/pull/570

This minor release fixes a couple of issues found in the build system after the major release 0.7 of HiOp.

What's Changed

• fixing memory leaks in cuSolver by @kswirydo in https://github.com/LLNL/hiop/pull/554
• moves pnnl_cleanup to .pre stage and removes files based on age by @jdmacam in https://github.com/LLNL/hiop/pull/555
• Condensed/normaleqn linear system without CUDA/RAJA by @nychiang in https://github.com/LLNL/hiop/pull/557
• Fix mds raja example for raja without GPU by @nychiang in https://github.com/LLNL/hiop/pull/558
• git submodule fixes by @cnpetra in https://github.com/LLNL/hiop/pull/561
• Misc build system fixes related to HIOP_SPARSE by @cnpetra in https://github.com/LLNL/hiop/pull/560
• Disable openmp when raja is not used by @nychiang in https://github.com/LLNL/hiop/pull/562
• Pre stage pipeline fix by @ryandanehy in https://github.com/LLNL/hiop/pull/564
• New tag release - 0.7.1 by @cnpetra in https://github.com/LLNL/hiop/pull/565

New Contributors

• @jdmacam made their first contribution in https://github.com/LLNL/hiop/pull/555
• @ryandanehy made their first contribution in https://github.com/LLNL/hiop/pull/564

Full Changelog: https://github.com/LLNL/hiop/compare/v0.7.0...v0.7.1

• Fortran interface and examples
• Bug fixing for sparse device linear solvers
• Implementation of CUDA CSR matrices
• Iterative refinement within CUSOLVER linear solver class
• Improved robustness and performance of mixed dense-sparse solver for AMD/HIP

This tag provides an initial integration with ginko, fixes a couple of issues, and add options for (outer) iterative refinement.

This version/tag provides a workaround for an issue in the HIP BLAS and updates the RAJA code to better operate with the newer versions of RAJA.

The salient features of v0.6.0 are

• the release of the primal decomposition (PriDec) solver for structured two-stage problems
• improved support for (NVIDIA) GPUs for solving sparse optimization problems via NVIDIA's cuSOLVER API and newly developed condensed optimization kernels.

Other notable capabilities include

• improved accuracy in the computations of the search directions via Krylov-based iterative refinement
• design of a matrix interface for sparse matrices in compressed sparse row format and (capable) CPU reference implementation