Skip to main content

News

NERSC Summer Student Works Toward Machine Learning For a Rainy Day

James Duncan discusses his work at the 2022 summer student poster session. (Credit: Margie Wylie, Berkeley Lab)

LUX-ZEPLIN Data-Management Needs Fuel Team Approach

Data collection from the LZ detector began in 2021, and the data pipeline runs 24/7. As LZ’s U.S. data center, NERSC provides support for data movement, processing, and storage – to the tune of 1PB of data annually. The data comes to NERSC from the Sanford Underground Research Facility in South Dakota via ESnet.

Simulations Run at NERSC Target ‘Forever Chemicals’ in Soils

There are many ways PFAS can enter the environment, all of which increase the odds of finding these chemicals in our food or water. (Credit: Michigan Department of Environment, Great Lakes and Energy)

Neuroscience Simulations at NERSC Shed Light on Origins of Human Brain Recordings

Rendering of a full-scale biophysically detailed model of a cortical column. The simulation had ~30,000 neurons and ~30,000,000 compartments. The color indicates the flow of current through compartments. The visualization was created by Burlen Loring, Lawrence Berkeley National Laboratory. (Credit: Burlen Loring, Berkeley Lab)

Jeff Broughton Looks Back, and Prepares for the Next Thing

Jeff Broughton stands by the Edison supercomputer that he helped bring to NERSC, one of his proudest accomplishments during a long career.

Simulations at NERSC Drive Progress Toward Nuclear Fusion

The NSTX-U is a nearly spherical, toroidal tokamak, meaning its outer shell is nearly perfectly round, and the inside is shaped like a donut.

Machine Learning Fuels Materials Science and Search in Continuous Action Spaces

Two representations show the algorithm's effectiveness at predicting force fields for 54 elements across the periodic table.

Researchers Narrow Down Mass of Sought-After Axion Particle

In their search for the axion particle, scientists focus their simulations on axion strings, topological defects in the soup of the early universe. Here, an axion string is rendered as a dark blue loop emitting axions. (Credit: Malte Buschmann, Princeton University)

New Encoder-Decoder Overcomes Limitations in Scientific Machine Learning

Figure: A general depiction of the encoder-decoder framework with an integrated CRF-RNN layer. The input data is fed into a chosen encoder, which then is upsampled by the U-Net-based decoder. Both the decoder output and the initial image serve as inputs for the CRF-RNN layer, which produces the final pixel-based prediction. (Credit: M. Avaylon, T. Perciano, Z. Bai)

San Francisco and Berkeley Lab Team Up on Pioneering Climate Study

Running simulations at NERSC, the research collaboration found that the effect of climate change on future storms in the San Francisco Bay Area will be significant, leading to more powerful storms unleashing substantially more water. (Credit: Brocken Inaglory via Wikimedia Commons)

Hydrogen Fuel Cells: An Environmentally Friendly Alternative

Hydrogen Fuel Cell Electric Vehicles: How they Work Hydrogen fuel cell electric vehicles (FCEVs) are similar to electric vehicles in that they use an electric motor instead of an internal combustion engine to power the wheels. However, while electric vehicles run on batteries that must be plugged in to recharge, FCEVs generate their electricity onboard. In a fuel cell, hydrogen gas from the vehicle’s fuel tank combines with oxygen from the air to generate electricity with only water and heat as byproducts of the process. Fueling a hydrogen FCEV is similar to refilling a vehicle’s gas tank, but using a nozzle from a designated hydrogen dispenser at a public station. Refueling times are also similar: FCEVs can be refueled in as little as 5 minutes. Some FCEVs can go more than 300 miles on one tank of hydrogen fuel — greater than the distance from St. Louis to Chicago — and fuel economy close to 70 MPGe (miles per gasoline gallon equivalent). For more details about hydrogen FCEVs and how they work, visit this U.S. Environmental Protection Agency information site.

Berkeley Lab Computing Resources Enable Deeper Understanding of Supernovae Explosions

A dead star explodes as a Type Ia supernova. (Image credit: NASA)
Computing Sciences logo

Connect

Our Organization