Delivers a 3-part Spring 2024 FRACTALS Fellow Series
From present to future: networks for high-performance computing | Register
2024-02-20 | 2024-03-12 | 2024-03-26 (Noon Eastern)
Description
Networks for high-performance computing require low-latency and high-bandwidth to maintain an efficient system. These requirements can be achieved by networks whose underlying graph has a low-diameter and a large number of edge-disjoint spanning trees. A network topology, which is based on a graph, describes the arrangement of the elements in the network. The arrangement is crucial to guaranteeing the reliability of the network while minimizing latency and maximizing bandwidth.
In this series, we will survey network topologies for high-performance computing for the present and the future systems. We begin by introducing and exploring network topologies used for high-performance computing today. We’ll then focus on network topologies for high-performance computing for the future, with the opportunity to even design our own!
Bio
Aleyah Dawkins is a PhD candidate at George Mason University, graduating in May 2024. Her main research interests are in combinatorics, particularly graph theory. Aleyah is currently a guest student at Los Alamos National Laboratory (LANL) where she researches mathematical designs and properties of high-performance computing networks to be applied to exascale and post-exascale networks, in collaboration with researchers from Colgate University, ETH Zurich, Intel, and LANL. Her professional mission is to create inclusive learning environments to address retention of a diverse community in mathematics.