From Language to Silicon: Programming Systems for Sparse Accelerators

Explore a comprehensive lecture from the Allen School Colloquia Series where Olivia Hsu, a final-year Ph.D. candidate from Stanford University, discusses programming systems for sparse accelerators. Discover how modern hardware development is shifting toward domain-specific accelerator design due to technology scaling plateaus and ongoing performance demands. Learn about two innovative systems: the Sparse Abstract Machine (SAM), which introduces a unified abstract machine model and compiler abstraction for sparse dataflow accelerators, and Mosaic, which offers modular and portable compilation solutions for heterogeneous sparse accelerators. Understand the challenges and future directions in creating usable programming systems that can democratize access to specialized hardware acceleration across various domains. Olivia Hsu, an NSF Graduate Research Fellow and 2024 Rising Star in EECS whose research earned a distinguished paper award at PLDI 2023, shares insights from her work at the intersection of computer architecture, programming systems, compilers, and digital circuits.