ICL Research Profile

BONSAI

Overview

The goal of the BEAST OpeN Software Autotuning Infrastructure (BONSAI) project is to develop a software infrastructure for using scalable, parallel hybrid systems to carry out large, concurrent autotuning sweeps in order to dramatically accelerate the optimization process of computational kernels for GPU accelerators and many-core coprocessors. BONSAI is based on Python 2.7 and the Python-based LANguage for Autotuning Infrastructure (LANAI) syntax.

As the name suggests, BONSAI builds on the experiences of the Bench-testing Environment for Automated Software Tuning (BEAST) project, which prototyped and validated an autotuning workflow consisting of generation and pruning of the parameter search space; compilation, benchmarking, and profiling of the kernels that pass the pruning; and collection, analysis, and visualization of the performance data.

Find out more at https://bitbucket.org/icl/bonsai/

Project Handout