Topic: HPC Systems and Software

The 2019 Department of Energy (DOE) Performance, Portability and Productivity meeting is slated for April 2–4, 2019, in Denver, CO, where attendees will have the opportunity to share ideas and updates on performance portability.

In this video from the 2019 Stanford HPC Conference, LLNL's Todd Gamblin presents "Spack – A Package Manager for HPC."

In this video from the 2019 Stanford HPC Conference, LLNL's Rob Neely presents "Sierra – Science Unleashed." Sierra is NNSA’s first large-scale production heterogeneous system.

Umpire is a resource management library that allows the discovery, provision, and management of memory on next-generation architectures.

Profile of LLNL's Lori Diachin, who has over 25 years experience in applied mathematics research including mesh quality improvement, mesh component software, numerical methods, and parallel computing.

Computer scientist Greg Becker contributes to HPC research and development projects for LLNL’s Livermore Computing division.

Highlights include debris an shrapnel modeling at NIF, scalable algorithms for complex engineering systems, magnetic fusion simulation, and data placement optimization on GPUs.

Users need tools that address the bottlenecks of exascale machines, work seamlessly with the programming models on the target machines, scale to the full size of the machine, provide the necessary automatic analysis capabilities, and be flexible and modular enough to overcome the complexities and changing demands of exascale architectures.

This open-source file system framework supports hierarchical HPC storage systems by utilizing node-local burst buffers.

Highlights include CASC director Jeff Hittinger's vision for the center as well as recent work with PruneJuice DataRaceBench, Caliper, and SUNDIALS.

As supercomputing power increases, the cost of communicating data both on- and off-node has become a critical factor affecting the overall performance of a parallel application. Livermore's interconnection networks projects improve the communication and overall performance of parallel applications using interconnect topology-aware task mapping.

The PRUNERS Toolset offers four novel debugging and testing tools to assist programmers with detecting, remediating, and preventing errors in a coordinated manner.

Livermore’s archive leverages a hierarchical storage management application that runs on a cluster architecture that is user-friendly, extremely scalable, and lightning fast.

LLNL's Advanced Simulation Computing program formed the Advanced Architecture and Portability Specialists team to help LLNL code teams identify and implement optimal porting strategies.

BLT software supports HPC software development with built-in CMake macros for external libraries, code health checks, and unit testing.

Highlights include the latest work with RAJA, the Exascale Computing Project, algebraic multigrid preconditioners, and OpenMP.

Highlights include complex simulation codes, uncertainty quantification, discrete event simulation, and the Unify file system.

Highlights include recent LDRD projects, Livermore Tomography Tools, our work with the open-source software community, fault recovery, and CEED.

A new software model helps move million-line codes to various hardware architectures by automating data movement in unique ways.

Sphinx, an integrated parallel microbenchmark suite, consists of a harness for running performance tests and extensive tests of MPI, Pthreads and OpenMP.

“If applications don’t read and write files in an efficient manner,” system software developer Elsa Gonsiorowski warns, “entire systems can crash.”

Highlights include the HYPRE library, recent data science efforts, the IDEALS project, and the latest on the Exascale Computing Project.

Apollo, an auto-tuning extension of RAJA, improves performance portability in adaptive mesh refinement, multi-physics, and hydrodynamics codes via machine learning classifiers.

Large Linux data centers require flexible system management. At Livermore Computing, we are committed to supporting our Linux ecosystem at the high end of commodity computing.

Babel is a high-performance language interoperability tool. The project is mainly developed at the Center for Applied Scientific Computing (CASC) LLNL. Babel started as an internal Lab Directed Research and Development (LDRD) project in 2000 and has been under constant development since then. It is now funded mainly under the U.S. Department of Energy (DOE) Office of Science's SciDAC program.