On Oct 17, 2009, at 7:17 AM, Robert Coup wrote:
On Fri, Oct 16, 2009 at 7:06 PM, Sean Gillies
<[email protected]> wrote:
GIS data would be herded through processing steps by the framework,
which calls your specific code at the right point. It probably
allows you to chain processing steps, split features into several
streams, and merge them again. Maybe it should allow one to
distribute work among the machines of a network a la Hadoop or
octo.py.
I built a prototype a while back of doing GeoProcessing via
MapReduce. The use case was things that ArcGIS sucks at: running a
set of relatively simple "steps" over a large dataset (for those who
know ArcGIS, it tends to grind away for many hours/days doing stuff,
then crashes). The specific use case was figuring out the walking
distance between every property in a city and the nearest park.
The idea was to define relatively simple MapReduce processing steps
in Javascript/Python. Then drop them into Hadoop, let it distribute
each step, and have the steps chained together.
Rob :)
Did your prototype suggest what a framework would need, say if you
wanted to plug in arbitrary "walk score" algorithms, or compute for
cycling instead?
--
Sean
