Jiuyang Liang’s computer screen flickers with a million atoms in motion—lithium ions darting through a dense sea of TFSI anions, each sphere a proxy for the invisible dance powering next-generation batteries. This simulation, once a weeks-long computational marathon, now unfolds in days, thanks to a breakthrough from the Flatiron Institute that’s rewriting the rules of molecular dynamics. More than 20% of the world’s top 500 supercomputers are dedicated to such simulations, which underpin everything from drug discovery to materials science. Now, those simulations can run 2.5 to seven times faster—without losing a shred of accuracy.
The secret lies not in faster chips or bigger data centers, but in smarter math. Shidong Jiang and his team at the Flatiron Institute’s Center for Computational Mathematics have resurrected and refined a classical mathematical function to dramatically reduce the computational load of calculating electrostatic forces between atoms. In the most widely used molecular dynamics software, GROMACS, their method delivers a fivefold speedup at high accuracy settings. That means simulations that once took 100 days can now finish in 20—slashing both time and energy use.
The implications ripple across science. For researchers like Pilar Cossio, who simulates proteins in aqueous environments, the advance means more questions answered in less time. Molecular dynamics simulations slice time into trillionths of a second to capture atomic vibrations, often requiring trillions of steps to model just milliseconds of real behavior. "Even with great hardware, we are limited to hundreds of nanoseconds a day," says Sonya Hanson, another Flatiron researcher. Now, that bottleneck is cracking.
What makes the method revolutionary is its simplicity and compatibility. It can be plugged into existing software workflows with minimal changes, opening the door to rapid adoption. No new supercomputers needed—just a smarter way to compute the forces that bind atoms. The team, including lead author Jiuyang Liang, software engineer Libin Lu, project leader Alex Barnett, and CCM Director Leslie Greengard, published their results in Nature Communications on May 21.
Anthony Costa, director of digital biology at Nvidia, calls the work a "testament to the importance of applied mathematical research," noting that the field had seen only incremental gains over the past decade. As simulations grow larger and more complex—from battery electrolytes to folding proteins—efficiency becomes as critical as accuracy. With this leap, science doesn’t just run faster. It scales smarter, poised to unlock discoveries once too slow, too costly, or too energy-intensive to pursue.