Nov. 20, 2020
The HPC world, particularly in the United States, is waiting for the next series of transitions to far larger machines with exascale capabilities. By this time next year, the bi-annual ranking of the top 500 most powerful systems will be refreshed at the top as Frontier, El Capitan, Aurora and other Department of Energy supers come online.
In the meantime, there have been several pre-exascale systems that appeared more recently across the U.S. national labs, not to mention several research workhorses that sported true architectural diversity, from Arm-based processors, to the latest GPUs from both Nvidia and AMD, not to mention AMD as a rising CPU force in HPC or the new wave of AI-specific accelerators.
Lawrence Livermore National Lab has been at the forefront of all of those architectures over the last year in particular across systems like the just-announced Mammoth machine, the expanded Corona supercomputer, the existing Sierra and Lassen supercomputers that are paving the way for El Capitan, and as of today, yet another machine in the lineup, Ruby, a top 100-class all CPU (Intel Platinum) super.
When one of Hye-Sook Park’s experiments goes well, everyone nearby knows. “We can hear Hye-Sook screaming,” she’s heard colleagues say.
It’s no surprise that she can’t contain her excitement. She’s getting a closeup look at the physics of exploding stars, or supernovae, a phenomenon so immense that its power is difficult to put into words.
Rather than studying these explosions from a distance through telescopes, Park, a physicist at Lawrence Livermore National Laboratory, creates something akin to these paroxysmal blasts using the world’s highest-energy lasers.
About 10 years ago, Park and colleagues embarked on a quest to understand a fascinating and poorly understood feature of supernovas: shock waves that form in the wake of the explosions can boost particles, such as protons and electrons, to extreme energies.
Membrane-based systems have great potential as low-energy alternatives in applications like desalination, pharmaceutical recovery, purification and waste treatment.
Scientists at Lawrence Livermore have created the largest defect-free membranes reported to date that fully exploit the unique mass transport properties of carbon nanotubes as flow channels.
Chemically robust materials that selectively transport molecules at a high rate are key for developing advanced membrane systems outperforming state-of-the-art products. Carbon nanotubes – all-carbon channels more than 50,000 times thinner than a human hair – belong to this class of highly promising membrane building blocks. Unlike conventional porous materials, these tiny channels allow for exceptionally fast gas and liquid flow that gets even faster as the tubes get smaller.
To reap the most benefits of these materials, maximizing the density of open carbon nanotubes across the membrane is critical. Researchers at LLNL grew high density, single-walled nanotubes on 4-inch silicon wafers and used them to create membranes with exceptional transport properties at scale.
On a cosmic timescale, 10 years or even a hundred years is the blink of an eye. Cosmic timescales are measured in millions or billions of years. Scientists from the Lawrence Livermore National Laboratory recently reported that research indicates our solar system formed in less than 200,000 years.
Researchers came to this conclusion after looking at isotopes of the element molybdenum found in meteorites. The material makes up the sun and the rest of the solar system and came from the collapse of a large gas and dust cloud about 4.5 billion years ago. By observing other solar systems that formed similarly to our solar system, astronomers were able to estimate that it likely took about 1 to 2 million years for the collapse of the cloud and the ignition of a star.
Before this study, the solar system formation timeframe was not clearly known. The new research shows the gaseous cloud collapse that led to the formation of the solar system happened very quickly, in less than 200,000 years. To put this timeframe in a better perspective, scientists on the project said that if we scale this to compare to human pregnancy, the pregnancy would last about 12 hours rather than nine months.
Lawrence Livermore can lay claim to housing four of the world’s 100 most powerful supercomputers, more than any other institution according to the TOP500 List announced earlier this week during the virtual Supercomputing 2020 conference (SC20).
The 125-petaFLOP peak Sierra, the National Nuclear Security Administration’s flagship supercomputer, remained third in the world, achieving 94.6 petaFLOPs (floating operations per second) on the High Performance LINPACK (HPL) benchmark. The LINPACK reflects the performance of a dedicated system for solving a dense system of linear equations and is used to determine the Top500.
Sierra ranked behind the reigning No. 1, Japan’s Fugaku supercomputer — which upped its HPL performance to 442 petaFLOPs from June’s TOP500 list — and Oak Ridge National Laboratory’s IBM/NVIDIA powerhouse Summit, which remained the fastest system in the United States at 148.8 petaFLOPs. Sierra also was 15th on the Green500 List of the world’s most energy-efficient supercomputers.