Tal Ben-Nun named to Forbes Israel's "30 under 30"

Tal Ben-Nun is a post-doctoral researcher at the Scalable Parallel Computing Laboratory (SPCL), led by Professor Torsten Hoefler. He is leading a project that will revolutionize the way people program computers and supercomputers, using intersections between Machine Learning and High-Performance Computing.

After completing his studies and doctorate in Computer Science at the Hebrew University of Jerusalem, Tal moved to ETH Zurich. Since then, he has been doing postdoctoral research at the Department of Computer Science. Here he uses supercomputers for various purposes, from weather forecasts to artificial intelligence. He is also working to make programming more accessible to scientists and other people who use intensive computations in their work.

Scientists know how to mathematically describe their research problems and translate it into complex algorithms, but the challenge arises when it comes to translating the algorithms into efficient software. Performance is crucial because some of these programs run over very long periods, sometimes even weeks. Any improvement in their capabilities, therefore, means enormous time savings. Traditionally, talented students, researchers, and even software engineers are employed to program these kinds of software.

This is where Tal’s work comes in. His research at SPCL aims to show programmers new ways in which they can accelerate their software by separating the algorithm from computation. This allows scientific applications to be optimized with the click of a button.

Tal and Prof. Hoefler’s group at ETH Zurich are among the world's leading research groups in parallel and distributed systems, as well as distributed deep learning. They work with one of the fastest computers in the world – Piz Daint, right here in Switzerland – in order to enable the scientific applications of the future.