- Number 349 |
- October 31, 2011
Python snakes into Global Arrays Toolkit
Researchers at DOE’s Pacific
Northwest National Laboratory
expanded the Global Arrays
Toolkit by integrating in the
programming language Python,
making it easier for programmers
to write codes and incorporate
features.
While many of us fear large reptiles, materials scientists and other researchers embrace Python—the easy-to-use programming language, not the snake. At DOE’s Pacific Northwest National Laboratory, computer programmers have included full support for Python in the Global Arrays Toolkit. The toolkit makes programming on distributed memory computers as easy as using shared memory on a desktop and scales to today’s top supercomputers. This allows for the simulation of larger, complex systems such as in computational chemistry, materials science, and computational fluid dynamics which impact national energy use.
Using the Global Arrays Toolkit results in a faster time to solution and a better understanding of the data and processes being evaluated. Integrating Python allows programmers to easily customize the Global Arrays Toolkit when they need shared memory for a distributed memory computers, improving and expanding researchers' ability to access the necessary data. A portion of this work was performed on EMSL's Chinook supercomputer. This work was funded by the DOE's Basic Energy Sciences.
[Kristin Manke, 509.372.6011,
kristin.manke@pnnl.gov]
