People & Programs
Dan Stanzione
Letter From the Director
TACC Executive Director Dan Stanzione reflects on past year and new developments
While sitting in a recent meeting of users of the Frontera system, I was struck by the incredible breadth of the work for which researchers use TACC computers.
What struck me is the wide range of time and length scales that different experiments use. Researchers doing work in materials were using timesteps in their simulations of picoseconds—trillionths of a second. Other researchers working on cosmology were simulating a range of Gigayears—billions of years. There are roughly 30 octillion (yes, that’s a real number, a 1 with 27 zeroes after it) picoseconds in a Gigayear.
There is a similar range in length where the measurements move from angstroms to megaparsecs (a 10 billionth of a meter to ~3 million light years). There are roughly 300 nonillion angstroms in a megaparsec (also a real number, a 1 followed by 30 zeroes).
It’s hard to conceive of any problem in the Universe that doesn’t fit somewhere within these bounds. The numbers speak to the power of computing as a universal instrument of engineering and science, capable of tackling an extraordinary range of challenges. They also speak to our unquenchable thirst for more performance from supercomputers, and why we push from petaflops to exaflops and beyond.
As we head towards the close of 2024, TACC is taking the next leap forward in pushing computing towards these vast scales. In July 2024, after a planning process of many years, TACC was awarded the contract to construct the U.S. National Science Foundation Leadership-Class Computing Facility (NSF LCCF). TACC and our partners across the nation will not only deploy the next generation of supercomputing technologies that are an order of magnitude faster than we have ever had before, but we will also make the LCCF part of NSF’s collection of long-running facilities, putting computing on equal footing with other large scientific instruments from telescopes to particle accelerators to neutrino detectors.
This new facility will not only mean more computing, but a better ability to engage in long-term stable partnerships with the rest of the scientific community. It will take two years to bring all aspects of the NSF LCCF to life from construction of new data centers to the installation and testing of new computing and storage systems.
But we haven’t paused in the meantime. During this past year TACC brought the new Stampede3 system and the AI-focused Vista system online to continue to provide our users with state-of-the-art infrastructure for research.
Early last year we added our 100,000th user account—meaning 100,000 different researchers have now relied on TACC for at least some of their computational needs.
The award of the NSF LCCF is a big milestone for us, the largest contract in our history. Many challenges lie ahead from working with enormous power to new computational infrastructure demands to adapting to the lower-precision chips being produced for AI.
We look forward to collaborating with you to meet these challenges in the years to come!