Proceedings of the 11th Python in Science Conference
Lists
Only you see this
Abstract
We use a combination of code-generation, code lowering, and just-in-time compilation techniques called SEJITS (Selective Embedded JIT Specialization) to generate highly performant parallel code for Bag of Little Bootstraps (BLB), a statistical sampling algorithm that solves the same class of problems as general bootstrapping, but which parallelizes better.We do this by embedding a very small domain-specific language into Python for describing instances of the problem and using expert-created code generation strategies to generate code at runtime for a parallel multicore platform.The resulting code can sample gigabyte datasets with performance comparable to hand-tuned parallel code, achieving near-linear strong scaling on a 32-core CPU, yet the Python expression of a BLB problem instance remains source-and performance-portable across platforms.This work represents another case study in a growing list of algorithms we have "packaged" using SEJITS in order to make high-performance implementations of the algorithms available to Python programmers across diverse platforms.
BibTeX
@inproceedings{varoquaux2012proceedings,
author = {Gaël Varoquaux and Travis Vaught and Jarrod Millman and Terrel and Aron Ahmadia and Kaust Nathan Bell and John Hunter and Matthew Knepley and Matthew Terry and Matthew Turk and Peter Wang and Continuum Analytics and Dharhas Pothina and Chris Colbert and Inc Spotz and Aakash Prasad and David Howard and Shoaib Kamil and Armando Fox and Alejandro Weinstein and Michael Wakin and Cyrus Harrison and Harinarayan Krishnan and Geoffrey Poore and Self-Driving Lego and Mindstorms Robot and Jacob Barhak and Jacob Frelinger and Adam Richards and Cliburn Chan and Matthew Rocklin and Robert Johansson and Paul Nation and Mads Ruben and Burgdorff Kristensen and Simon Andreas and Frimann Lund and Troels Blum and Brian Vinter and Tom Bertalan and Akand Islam and Roger Sidje and Eric Carlson and Adam Hughes and Esther Lederberg and John Hall and John Hasbrouck and Van Vleck and Andre Geim and Clinton Davisson and Robert Richardson and Gabriel Lippmann Gang and Polykarp Kusch and Rob Walker and Hendrik Lorentz and Pierre-Gilles De Gennes and Henri Becquerel and Jack Kilby and William Cullen and Bryant Sydney and Shelton and Steven Weinberg and Steven Landsburg and John Kennedy and John Keese and E White and Snow Chen and Ning Yang and Itiro Tomonaga and Paaras Troy and Robert Serber and Frits Zernike and Albert Michelson and Abdus Salam and Douglas Osheroff and Cheshire and Maclean' and Plymouth and Rudolf Mssbauer and Kai Siegbahn and Bertram Brockhouse and Wolfgang Pauli and Edward Appleton and Riccardo Giacconi and Max Born and Sasha Frere-Jones and Wnet Henry and Way Kendall and Donald Graham and Norman Ramsey and Joseph Taylor and Hagen Kleinert and Charles Kao and El Diario and La Prensa and Niels Bohr and Ernst Ruska and William Shockley and General Slocum and Thomas Jefferson},
title = {{Proceedings of the 11th Python in Science Conference}},
booktitle = {Proceedings of the Python in Science Conferences},
year = {2012},
doi = {10.25080/majora-54c7f2c8-00d},
}Loading related content...