Topic: HPC Systems and Software

The Summit Sierra team, consisting of 45 staff at LLNL and ORNL, received a DOE Secretary’s Achievement Award for delivering, respectively, Sierra and Summit supercomputers.

The extreme-scale scientific software development kit (xSDK) is an ecosystem of independently developed math libraries and scientific domain components.

A software product from the ECP called UnifyFS can provide I/O performance portability for applications, enabling them to use distributed in-system storage and the parallel file system.

At SC19, there were Spack events each day of the conference. Spack is an open-source scientific software package manager for HPC environments, MacOS, and Linux platforms.

The 2019 International Conference for High Performance Computing, Networking, Storage, and Analysis—SC19—returned to Denver. Once again LLNL made its presence known as a force in supercomputing.

Computing’s 22nd hackathon was held on October 24–25. The event has become so popular that a fourth hackathon will be added to the seasonal rotation in early 2020.

The HPCwire Editors’ and Readers’ Choice awards for Top Supercomputing Achievement recognized Cray, LLNL, and two other labs for developing the first U.S. exascale-class supercomputers.

Penguin Computing announced that Corona, a high performance computing cluster delivered to LLNL in 2018, has been upgraded with the newest AMD Radeon Instinct MI60 accelerators.

The latest TOP500 List of the world's most powerful computers was released at the International Supercomputing Conference for HPC, Networking, Storage and Analysis (SC19) in Denver.

Penguin Computing, along with its partners Intel and CoolIT, has shipped LLNL's Linux-based cluster, known as “Magma,” to the Laboratory.

Three LLNL teams, including the Spack package manager, netted regional awards for technology transfer from the Federal Laboratory Consortium (FLC) last week.

Of the Lab's four R&D 100 Award winners for 2019, two are open-source software projects: SCR and Spack.

In the third of three news features, LLNL-developed SCR software will contribute to the efficiency and effectiveness of the HPC leadership-class systems of tomorrow.

The Taurus Environmental Information Management System, or TEIMS, system manages collaborative tasks, site characterization, modeling, risk assessment, decision support, operations tracking, compliance monitoring, and regulatory reporting for the Environmental Restoration Department.

LLNL’s Computing Directorate will descend upon the 2019 Supercomputing Conference. Find us at tutorials, workshops, panels, poster/paper sessions (with a Best Paper finalist), and the Job Fair.

In the second of three news features, LLNL-developed OpenZFS software will contribute to the efficiency and effectiveness of the HPC leadership-class systems of tomorrow.

In the first of three news features, LLNL-developed Flux software will contribute to the efficiency and effectiveness of the HPC leadership-class systems of tomorrow.

After years of preparation, LLNL’s upgraded Ares code runs a 98-billion-element simulation on 16,384 GPUs on the Sierra supercomputer.

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

LLNL heads to the 31st annual Supercomputing Conference (SC19) in Denver, Colorado, on November 17–22, 2019.

Cray, the DOE’s Exascale Computing Project, the DOE labs that will house the nation’s first three exascale supercomputers have established National Exascale Day to be celebrated annually (Oct. 18).

Computational Scientist Ramesh Pankajakshan came to LLNL in 2016 directly from the University of Tennessee at Chattanooga. But unlike most recent hires from universities, he switched from research professor to professional researcher.

LLNL's Lori Diachin, deputy director of the DOE's Exascale Computing Project, was one of three authorities on the topic of exascale computing to be interviewed and featured on builtin.com.

The NNSA’s first exascale supercomputer, El Capitan, will have a peak performance of more than 1.5 exaflops (1.5 quintillion calculations per second) and an anticipated delivery in late 2022.

Highlights include perspectives on machine learning and artificial intelligence in science, data driven models, autonomous vehicle operations, and the OpenMP standard 5.0.