Stellar: A DSL to Build and Explore Sparse Accelerators
Sparse accelerators have traditionally been designed in a largely ad-hoc manner, which makes both comparisons between existing accelerators, and the design space exploration of future accelerators, more difficult. This ad-hoc design process is partly caused by the lack of simple and expressive abstractions that can accurately describe all the different components of sparse accelerator design. To address this need, we introduce a new DSL, called “Stellar,” which accurately describes and decouples (i) the functionality of a sparse accelerator, (ii) the scheduling of operations on it, (iii) the sparse data structures it supports, and (iv) the load-balancing techniques it supports. Stellar enables users to quickly design and explore sparse accelerators, and also generates both Verilog implementations and programming stacks for them.
Sat 17 JunDisplayed time zone: Eastern Time (US & Canada) change
11:20 - 12:30 | |||
11:20 10mTalk | Tags: A Framework for Distributed Event Ordering PLARCH Paul Mure Stanford University, Nathan Zhang Stanford University, Caroline Trippel Stanford University, Kunle Olukotun Stanford University | ||
11:30 15mTalk | Stellar: A DSL to Build and Explore Sparse Accelerators PLARCH Hasan Genc UC Berkeley, Hansung Kim University of California, Berkeley, Prashanth Ganesh University of California, Berkeley, Yakun Sophia Shao University of California, Berkeley | ||
11:45 15mTalk | PEak: A Single Source of Truth for Hardware Design and Verification PLARCH Caleb Donovick Stanford University, Ross Daly Stanford University, USA, Jackson Melchert Stanford University, Leonard Truong Stanford University, Priyanka Raina Stanford University, Pat Hanrahan Stanford University, USA, Clark Barrett Stanford University | ||
12:00 10mTalk | Challenges with Hardware-Software Co-design for Sparse Machine Learning on Streaming Dataflow PLARCH Rubens Lacouture Stanford University, Olivia Hsu Stanford University, Kunle Olukotun Stanford University, Fredrik Kjolstad Stanford University |