Hosted By CSPAC

Speaker: David Krasowska

Title: Architecting Task-Based Execution Across GPU Clusters and PIM Architectures

Abstract: Writing parallel and distributed programs is notoriously difficult. This talk explores how task-based programming frameworks can reduce this burden across increasingly heterogeneous and distributed hardware. First, I present cuNumeric.jl and Legate.jl, which transparently scale idiomatic Julia array workloads across CPUs and GPUs. Second, I shift focus to extending task-based programming models to emerging Processing-in-Memory (PIM) architectures. I introduce libvectordpu, a high-performance tasking model for direct DPU programming, and Legion-PIM, which integrates DPUs into the core Legion programming system. Ultimately, moving computation near memory requires more than just a new backend- it fundamentally reshapes our assumptions about task granularity, data partitioning, and locality.

---

What is CSPAC?

We are the CS PhD Advisory Council. We are a PhD student-led organization, and our mandate is to interface between PhD students and faculty on academic issues. We want to advocate for PhD students in the department, so if there is some way we can support you, please come talk to us. We welcome PhD students to our weekly meetings on Tuesdays, 4:00-5:00pm in Mudd 3501 and on zoom. We also welcome anonymous concerns/feedback at any time via this form. Anyone in the community can reach us at cspac@u.northwestern.edu.