Towards an Approximation-Aware Computational Workflow Framework for Accelerating Large-Scale Discovery Tasks
- PODC 2022
Michael Johnston is Research Scientist and Manager of the Next Generation Systems team at IBM Research Ireland. His background is in High Performance Computing, application design, and computational biophysics and biochemistry.
He has worked closely with the Hartree Centre, IBM’s collaboration with the UK’s Science and Technology Facilities Council (STFC), on a range of projects. This included the Square Kilometre Array (SKA) Telescope, aimed at studying the universe in unprecedented detail, and Oasis — a tool for estimating the cost of catastrophes. His main focus was on Materials Discovery, and from 2015-2020, he led a team of IBM researchers on the topic of virtual experiments which provide frictionless access to complex analytics techniques.
Recently he has been leading research into the topic of Discovery Systems, systems that aim to accelerate the scientific process, working with his team and worldwide collaborators on composable systems, runtimes for AI surrogate for physical models, and automation and optimization of the discovery process. He holds a PhD. in Computational Biophysics from Universitat Pompeu Fabra in Barcelona.