The System Tools and Advanced Runtimes (STAR) group of the Barcelona Supercomputing Center-Centro Nacional de Supercomputación (BSC-CNS) has announced the release of the STAR software stack (version 2024.05), featuring several updates and enhancements.
The 2024.05 release of the OmpSs-2 programming model introduces new features such as the Directory/Cache (D/C) for CUDA devices in Nanos6, alongside multiple enhancements to the nOS-V tasking library, and overall performance and bug fixes. Additionally, the enhanced LLVM/OpenMP runtime (libomp-v) now supports OpenMP free-agents and detailed instrumentation through ovni. This release discontinues support for the Mercurium compiler.
Key Updates in the 2024.05 Release:
Nanos6:
- Introduction of directory/cache (D/C) for CUDA devices.
- New device memory allocation API for D/C-managed memory.
- Enhancements to the ovni instrumentation for improved monitoring and analysis.
nOS-V:
- New task batch submission API that to optimize submission efficiency.
- Native support for mutex (nosv_mutex_t) and barriers (nosv_barrier_t)
- Activation of the turbo mode configuration (i.e., fast-math) by default, enhancing performance floating-point performance on x86-64 and aarch64 platforms.
LLVM/OpenMP (libompv):
- Implementation of the OpenMP free-agents feature and ovni instrumentation, enhancing flexibility and monitoring capabilities.
LLVM/Clang:
- Introduction of the OPENMP_RUNTIME environment variable to select the preferred runtime library.
- Several bug fixes and improvements to enhance stability and performance.
Ovni:
- Add support for OpenMP model to instrument the libompv runtime
- Updates to support documentation and configuration files for better usability.
- Other bugfixes and improvements
For more information and access to the repositories, please visit: https://github.com/bsc-pm/ompss-2-releases
Recent publications:
- D. Álvarez, Kevin Sala and Vicenç Beltran, “nOS-V: Co-Executing HPC Applications Using System-Wide Task Scheduling,” 38th IEEE International Parallel & Distributed Processing Symposium (IPDPS), San Francisco, California, USA, May 27-31, 2024
- K. Sala et al, “ALPI: Enhancing Portability and Interoperability of Task-Aware Libraries,” Asynchronous Many-Task Systems and Applications (WAMPTA), Knoxville, Tennessee, USA, Feb. 14-16, 2024
Funding and Acknowledges:
The STAR research group is co-financed by the Barcelona Zettaescale Laboratory, which receives funding from the Ministry for Digital Transformation and Public Services within the framework of the Resilience and Recovery Facility. Additionally, the STAR research group has been funded by the European Union’s Horizon 2020/EuroHPC research and innovation program under grant agreements No. 101034126 (EuPILOT), No. 955606 (DEEP-SEA), and No. 956416 (ExaFOAM). Additionally, projects PCI2021-122090-2A, PCI2021-121958, and PCI2021-121961 have been supported by the MCIN/AEI/10.13039/501100011033 and the European Union’s NextGenerationEU/PRTR. This work has also been supported by the Ramón y Cajal program [grant number RYC2019-027592-I] and the Severo Ochoa Centre of Excellence accreditation [grant number CEX2021-001148-S], both financed by the MCIN/AEI/10.13039/501100011033. Furthermore, the Departament de Recerca i Universitats de la Generalitat de Catalunya has funded the Programming Models research group at BSC-UPC [grant number 2021 SGR01007].
