supercomputer News

Supercomputer Research Sheds Light on the Death of Stars

Supernovae exhibit the most-energetic explosions, dispersing elements that make life possible into the universe. However, the energy source for the violent death of these massive stars is not known. Researchers using UT’s Kraken supercomputer have created three-dimensional simulations that have made great strides in uncovering the source.

New Supercomputing Method Helps Energy and Materials Research

Research being done on the supercomputer Kraken holds promise for overcoming limitations in the study of energy and materials applications. The method employs quantum mechanics to understand how nuclear effects change the dynamics of microscopic-size materials.

Supercomputer Work Makes Stride in Biofuels Research

Cellulase enzymes found in nature from sources such as wood-degrading fungi or in cows’ stomach compartments form one of the key catalysts for breaking down plant biomass to make biofuels. But, they remain quite expensive. Compute allocations from the Extreme Science and Engineering Discovery Environment (XSEDE) have made a breakthrough possible that could have big cost implications.

UT Network Infrastructure to Upgrade to 100G

The UT–Oak Ridge National Laboratory Joint Institute for Computational Sciences—and UT’s Office of Information Technology—have announced final plans to upgrade the bandwidth of UT’s wide area network for research and education to 100 gigabit per second (100G) capability by July 2014. This project makes UT an early adopter of the technology and will improve a wide range of big data and other science data flows.

NICS Project Aims to Piece Together Pathology Puzzle

As disease progresses over space and time in the body, high-resolution imaging can capture the changes taking place down to the sub-cellular level; meanwhile, huge sets of hereditary (genomic) information hold clues about the dynamics of illness. Comparing certain characteristics in the images with genomic and clinical data may be key in predicting disease progression and in targeting new treatments. The current work of a research team at UT’s National Institute for Computational Sciences revolves around making those very connections.

Professor Helps Design Software for the Next Generation of Supercomputer

Jack Dongarra, distinguished professor of computer science at UT is designing software that will be critical in making the next generation of supercomputers operational. For decades, supercomputers have been tackling the world’s most pressing challenges, from sequencing the human genome to predicting climate changes. But their power is limited and thus, so is our knowledge.

Professor Jack Dongarra Announces New Supercomputer Benchmark

The way the power of supercomputers is measured is about to change. Since 1993, Jack Dongarra, distinguished professor of computer science at UT has led the ranking of the world’s top 500 supercomputers. The much-celebrated bi-annual TOP500 list is compiled using Dongarra’s benchmark system, called Linpack. But Dongarra says Linpack hasn’t kept pace with supercomputing needs and must be updated.