Section: New Results

On Fusing Recursive Traversals of K-d Trees

Participants : Samyam Rajbhandari [OSU, USA] , Jinsung Kim [OSU, USA] , Sriram Krishnamoorthy [PNNL, USA] , Louis-Noel Pouchet [CSU, USA] , Fabrice Rastello, Robert J. Harrison [Stony Brook, USA] , P. Sadayappan [OSU, USA] .

Loop fusion is a key program transformation for data locality optimization that is implemented in production compilers. But optimizing compilers for imperative languages currently cannot exploit fusion opportunities across a set of recursive tree traversal computations with producer-consumer relationships. In this work, we develop a compile-time approach to dependence characterization and program transformation to enable fusion across recursively specified traversals over k-d trees. We present the FuseT source-to-source code transformation framework to automatically generate fused composite recursive operators from an input program containing a sequence of primitive recursive operators. We use our framework to implement fused operators for MADNESS, Multiresolution Adaptive Numerical Environment for Scientific Simulation. We show that locality optimization through fusion can offer significant performance improvement.

This work is the fruit of the collaboration 8.4 with OSU. The specific work on FuseT has been presented to the international conference CC 2016 [32] and the more general work on the improvement of MADNESS at the ACM/IEEE international conference SC 2016 [20].