Colloquium with Milinda Fernando (2026-01-23)

"Scalable Algorithms for Exascale Science and Engineering"

Milinda Fernando
University of Texas at Austin

Friday, Jan 23, 2026

Colloquium - 499 DSL Seminar Room: 03:30 to 04:30 PM Eastern Time (US and Canada)

In-person attendance is requested.

499 DSL Seminar Room

Zoom access is intended for external (non-departmental) participants only.

Click Here to Join via Zoom
Meeting # 942 7359 5552

Zoom Meeting # 942 7359 5552

Abstract:

In this talk, I will present my research on the development of scalable, hardware-aware numerical algorithms for solving complex partial differential equations (PDEs) on leadership-class supercomputers. By co-designing algorithms with architectural constraints in mind, my work enables high-fidelity simulations that were previously computationally infeasible. I will highlight key algorithmic advances and demonstrate their impact through three representative application areas:

(A) Micro-scale physics, where I develop scalable solvers for the Boltzmann transport equation to model electron transport in low temperature plasmas;
(B) Relativistic astrophysics, where communication-avoiding and extreme-scale AMR enables large-scale simulations of the Einstein field equations for binary black-hole mergers and resulting gravitational waves; and
(C) Real-time Bayesian inversion, where the underlying problem structure is exploited to develop fast Hessian matrix–vector products that enable extreme-scale, real-time Bayesian inversion for a tsunami early warning system.

Together, these examples illustrate a unifying theme: carefully designed scalable algorithms are a prerequisite for extracting scientific insight from next-generation computing platforms. I will conclude by outlining my vision for future research in extreme-scale algorithms and their role in advancing science and engineering in the exascale era.