The parallel cartographic modeling language (PCML) is a multi-institutional collaborative project aiming to create a computing language for cyberGIScientists that is designed for (1) usability, (2) programmability, and (3) scalability. PCML provides multi-core parallel processing for spatial operations while hiding the implementation complexities of parallelism.
Note PCML has been making progress behind-the-scenes. One of the primary developers (Eric Shook), moved to the University of Minnesota, which disrupted some of the public-facing PCML development. However, several students and Dr. Shook continue to advance PCML behind this Github repository so stay tuned for some very exciting updates very soon.
PCML has been used in several graduate and undergraduate courses taught at Kent State University and the University of Minnesota including 'Introduction to CyberGIS' and 'Advanced Geocomputing.' It was featured in a workshop (see the 30 minutes PCML demo below). We also have plans in place to teach it in several courses this Spring 2017 semester at different institutions.
PCML is easy to use for parallel spatial data processing. First, users import PCML. The second line tells PCML to use 4 processing cores for parallel processing. Next, the users will read an ASCII grid file and a GeoTIFF file as layer1 and layer2, respectively. The two layers are added together using a LocalSum raster operation. The output output layer (layer_out) is printed to the screen.
from pcml import *
PCMLConfig.num_procs = 4 # Run computation in 4 processes (default)
layer1 = ReadASCIIGrid("layer1.asc")
layer2 = ReadGeoTIFF("layer2.tiff")
layer_out = layer1 + layer2
layer_out.print_data()
Please also see test.py or examples.py for additional examples.
For detailed instructions please see https://github.com/HPCGISLab/pcml/blob/master/demo_30min_install.pdf
The Anaconda Python distribution package is available at http://continuum.io/downloads. It provides Python and several Python packages, but can be skipped if you have Python already.
Navigate to this page (i.e., the PCML page on GitHub at https://github.com/hpcgislab/pcml)
Scroll up and click the "Download ZIP" button on the right side.
A file named “PCML-master.zip” will be downloaded to your computer. Unzip it.
Open 'demo_30min.py' by clicking File->Open in the Spyder editor (included in the Ananconda package)
Open 'demo_30min_walkthrough.pdf' and follow the 30 minute tutorial.
Success!
Let us know! We are trying to make PCML as easy-to-use and as useful as possible. We live to collaborate and work on interesting problems (and we will debug your PCML code for you).
Submit a 'New issue' on github (https://github.com/HPCGISLab/pcml/issues) or email Eric Shook (eshook@gmail.com).