Topic: Data Movement and Memory

A unique laser optic design and two novel open-source software projects bring the Laboratory’s R&D 100 awards total to 182.

This open-source library automatically maps HPC applications to supercomputers efficiently and transparently, so users can run on Day One on any system.

Research recognized at the HiPC IEEE conference proposes using an optimized version of OpenMP for vendor-agnostic GPU performance, portability, and scalability.

The latest issue of R&D World's magazine showcases LLNL's 2024 winning technologies, including UnifyFS and UMap software projects.

Specialized hardware modules and software libraries to optimize memory access while simultaneously increasing memory capacity for data-intensive applications.

This issue highlights some of CASC’s contributions to the DOE's Exascale Computing Project.

UMap uniquely exploits the prominent role of complex memories in today’s servers and offers new capabilities to directly access large memory-mapped datasets.

LLNL researchers have achieved a milestone in accelerating and adding features to complex multiphysics simulations run on GPUs, a development that could advance HPC and engineering.

PERM is a 'C' library for persistent heap management and is intended for use with a dynamic-memory allocator (e.g. malloc, free).

LLNL is participating in the 35th annual Supercomputing Conference (SC23), which will be held both virtually and in Denver on November 12–17, 2023.

The Data and Visualization efforts in the DOE’s Exascale Computing Project provide an ecosystem of capabilities for data management, analysis, lossy compression, and visualization.

Splitting memory resources in high performance computing between local nodes and a larger shared remote pool can help better support diverse applications.

LLNL participates in the ISC High Performance Conference (ISC23) on May 21–25.

An LLNL Distinguished Member of Technical Staff, Gokhale is considered an expert in her field, and continues to enjoy the fast pace of innovation and change in computing.

LLNL is participating in the 34th annual Supercomputing Conference (SC22), which will be held both virtually and in Dallas on November 13–18, 2022.

Winning the best paper award at PacificVis 2022, a research team has developed a resolution-precision-adaptive representation technique that reduces mesh sizes, thereby reducing the memory and storage footprints of large scientific datasets.

The MAPP incorporates multiple software packages into one integrated code so that multiphysics simulation codes can perform at scale on present and future supercomputers.

A Livermore-developed programming approach helps software to run on different platforms without major disruption to the source code.

Highlights include response to the COVID-19 pandemic, high-order matrix-free algorithms, and managing memory spaces.

Researchers develop innovative data representations and algorithms to provide faster, more efficient ways to preserve information encoded in data.

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

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