•  

Doubling the Performance of Python/NumPy with less than 100 SLOC

A session at Python for High Performance and Scientific Computing (PyHPC 2013)

Monday 18th November, 2013

10:30am to 11:10am (MST)

A very simple, and outside NumPy, commonly used trick of buffer-reuse is introduced to the NumPy library to speed up the performance of scientific applications in Python/NumPy. The implementation, which we name software victim-caching, is very simple. The code itself consists of less than 100 lines of code, and took less than one day to add to NumPy, though it should be noted that the programmer was very familiar with the inner workings of NumPy. The result is an improvement of as much as 2.29 times speedup, on average 1.32 times speedup across a benchmark suite of 15 applications, and at no time did the modification perform worse than unmodified NumPy.

About the speakers

This person is speaking at this event.
Simon A. F. Lund

Imagine that something short, concise and intriguing is written here... bio from Twitter

This person is speaking at this event.
Kenneth Skovhede

Post Doc at University of Copenhagen bio from LinkedIn

This person is speaking at this event.
Mads Ruben Burgdorff Kristensen

Postdoc at University of Copenhagen bio from LinkedIn

This person is speaking at this event.
Brian Vinter

Professor, Niels Bohr Institute at University of Copenhagen bio from LinkedIn

Next session in 505

11:10am ClusterShell, Python library and tools for scalable cluster administration by Henri Doreau, Aurélien Degrémont and Stéphane Thiell

Coverage of this session

Sign in to add slides, notes or videos to this session

Tell your friends!

When

Time 10:30am11:10am MST

Date Mon 18th November 2013

Session Hash Tag

#SC13

Short URL

lanyrd.com/sctbgq

Official event site

www.dlr.de/sc/pyhpc2013

View the schedule

Share

See something wrong?

Report an issue with this session