October 2, 2020
Scheduling transformations reorder a program’s operations to improve locality and parallelism. The polyhedral model is a general framework for composing and applying instance-wise scheduling transformations for loop-based programs, but there is no analogous framework for recursive programs. This work presents an approach for composing and applying scheduling transformations — like inlining, interchange, and code motion — to nested recursive programs. This talk will discuss the phases of the approach — representing dynamic instances, composing and applying transformations, and reasoning about correctness — and showing that these techniques can verify composed transformations’ soundness.
About Kirshanthan Sundararajah
Kirshanthan Sundararajah is a Ph.D. student in the School of Electrical and Computer Engineering at Purdue University advised by Milind Kulkarni. He is interested in Compilers, Programming Languages, and High-Performance Computing with an emphasis on optimizing irregular programs.