Today at Berkeley Lab

Supercomputers Help ID New Drug Leads to Fight Heart Disease

Using a unique computational approach to rapidly sample, in millisecond time intervals, proteins in their natural state of gyrating, bobbing, and weaving, a research team from UC San Diego and Monash University in Australia (using resources at NERSC) has identified promising drug leads that may selectively combat heart disease, from arrhythmias to cardiac failure. More>

You can leave a comment. View the Comment Policy.

Cori Supercomputer is Fifth on List of World’s Fastest Machines

When the 48th edition of the TOP500 list of the world’s top supercomputers was unveiled at SC16, NERSC’s Cori system debuted in the number five position with a Linpack rating of 14.0 petaflops. Go here for more on the Lab’s participation in SC16.

You can leave a comment. View the Comment Policy.

Celeste Enhancements Create New Opportunities in Sky Surveys

Celeste, a statistical analysis model designed to enhance sky surveys has gotten a major upgrade that dramatically scales up the process of cataloging astronomical objects. The Celeste code implements statistical inference to build a fully generative model to mathematically locate and characterize light sources in the sky. More>

You can leave a comment. View the Comment Policy.

Highly Capable Networking is Critical Component of Science Discovery

As scientific instruments grow more sensitive over the next few years, they are expected to produce data flows in excess of 1 terabyte per second. NERSC, ESnet and Cray are preparing for the onslaught with a project that uses software defined networking (SDN) to bring data to NERSC’s flagship supercomputer, Cori. More>

You can leave a comment. View the Comment Policy.

Multiscale HPC Captures Photovoltaics at the Nanoscale

Advances in ultrafast spectroscopy experiments and high performance computing are revolutionizing the ability to study the physical processes that occur in solar cells, down to the nanometer and femtosecond scale. When this data is combined with advanced calculations run on supercomputers like those at NERSC, the results bode well for the next generation of solar cells. More>

You can leave a comment. View the Comment Policy.

NERSC Issues ‘NESAP for Data’ Call for Proposals

NERSC is now accepting applications for participation in its new “NESAP for Data” program, an extension of the center’s popular NERSC Exascale Science Applications Program, a code optimization effort for next-generation supercomputers. Deadline for submissions is Nov. 1. More>

You can leave a comment. View the Comment Policy.

Edison Improves Performance, Energy Efficiency of Bioinformatics Application

A team of computer scientists and geneticists has demonstrated significant speedups of the epiSNP bioinformatics program using the Edison supercomputer at NERSC. The team used the application to perform pairwise comparisons of genotypes and how they affect traits such as height, weight or makeup of body tissues. More>

You can leave a comment. View the Comment Policy.

Toward Cost-Effective Polymer Electrolyte Fuel Cells

As the age of hybrid, electric and self-driving cars enters, interest remains high in finding the next generation of fuel cell technology. a research team running density functional theory calculations at NERSC has demonstrated how one type of fuel cell — polymer electrolyte fuel cells —can be made to run more efficiently and produced more cost-effectively. More>

You can leave a comment. View the Comment Policy.

Lab to Partner with U.S. Manufacturers on Five New HPC4Mfg Projects

The Department of Energy’s High Performance Computing for Manufacturing (HPC4Mfg) Program has announced $3.8 million in funding for 13 new industry projects in the second round of the HPC4Mfg program, and Berkeley Lab has been selected to partner with five of the new projects. More>

You can leave a comment. View the Comment Policy.

Joint Genome Institute Uncovers Earth’s Viral Diversity

JGI researchers, aided by the National Energy Research Scientific Computing Center (NERSC), used the largest collection of assembled metagenomic datasets to uncover over 125,000 partial and complete viral genomes, with infecting microbes in the majority. The findings, which were published in Nature, increase the number of known viral genes by a factor of 16. More>

You can leave a comment. View the Comment Policy.