3 papers
cs.NI2026
On Topology's Role in ML Training Performance
Sarah McClure, Tegan Wilson, Brad Karp +4
Modern machine learning training workloads run on large-scale networks of compute accelerators. The networks commonly deployed in these systems are typically variations of two basi…
cs.DS2026
Indirect Coflow Scheduling
Alexander Lindermayr, Kirk Pruhs, Andréa W. Richa +1
We consider routing in reconfigurable networks, which is also known as coflow scheduling in the literature. The algorithmic literature generally (perhaps implicitly) assumes that t…
cs.DS2025
Universal Connection Schedules for Reconfigurable Networking
Shaleen Baral, Robert Kleinberg, Sylvan Martin +3
Reconfigurable networks are a novel communication paradigm in which the pattern of connectivity between hosts varies rapidly over time. Prior theoretical work explored the inherent…