Scientists and engineers at Lawrence Livermore National Laboratory are among the top innovators throughout the globe according to a prestigious research and development award recognizing the top 100 innovations over the past year, with three projects out of the lab being recognized by what are dubbed the "Oscars of innovation."
R&D World Magazine recently announced this year's list of winners with LLNL being the birthplace of three of those projects and 179 awards total in the 61 years since R&D World's inception in 1978.
“The R&D 100 awards highlight the most game changing technologies each year,” LLNL Director Kim Budil said in a statement about the recognition. “Researchers at LLNL strive to address the most significant challenges facing the world today through innovative science and technology and these awards are an important recognition of the impact of this work.”
Two of the three projects out of the lab that are recognized this year include advances in supercomputers, with a third that leverages deep learning techniques to aid in treating and predicting outcomes for cancer patients.
Variorum is one of the former two projects, led by LLNL scientist Tapasya Patki and computer scientists Aniruddha Marathe, Barry Rountree, Eric Green, Kathleen Shoga and Stephanie Brink. It serves as a "vendor-neutral software library for exposing and monitoring the power, energy and performance of low-level hardware dials across diverse architectures in a user-friendly manner for supercomputers," according to LLNL officials.
Variorum is part of the U.S. Department of Energy's Argo project, serving as what lab officials call a "key component" of power management as part of a high-performance computing (HPC) initiative, and providing support for all three planned exascale supercomputers in the United States, including LLNL's El Capitan as well as parallel projects at other national laboratories and a number of other HPC systems.
The second innovation in supercomputing recognized is ZFP, an open-source software that functions by compressing numerical data and increases the speed of data movement – which has thus far been an obstacle to the advancement of HPC projects, as raw computer power has advanced over the years.
The project is led by LLNL computer scientists Danielle Asher and Mar C. Miller, with former LLNL employees Stephen Herbein, Matthew Larsen and Markus Salasoo having served on the team prior to their departures.
ZFP's major innovation is that it allows users to "cram larger data sets into memory or reduce how much disk space they need" as well as offering the ability to compress data before it is transmitted to different supercomputer nodes, thereby increasing the system's speed.
While supercomputers and the HPC world aren't part of most people's daily experience outside of national laboratories such as LLNL, the lab was recognized for its work along with four other national laboratories in developing the ECP-CANDLE project (Exascale Computing Project—CANcer Distributed Learning Environment).
The project leveraged resources from both the Department of Energy and the National Cancer Institute along with deep learning and machine learning techniques plus the lab's open source Livermore Big Artificial Neural Network toolkit with the goal of accelerating cancer research and treatment options.
LLNL's contribution to the project was led by computer scientist Brian Van Essen and Chief Computational Scientist Fred Streitz, with the current team including current LLNL employees Tom Benson, Adam Moody, Tal Ben-Nun, Nikoli Dryden, and Pier Fiedorowicz as well as former LLNL employees David Hysom, Sam Ade Jacobs, Naoya Maruyama and Tim Moon.
All three projects were supported by funding from the Laboratory Directed Research and Development program, which aims to facilitate "the undertaking of high-risk, potentially high-payoff projects at the forefront of science and technology," according to Lab officials.
The LLNL scientists and engineers in the three winning projects are set to be recognized along with workers on the 97 other projects throughout the globe to be named in this year's awards at a black-tie ceremony in San Diego on Thursday.