[SoC] SoC Proposal on a OGR Driver for MongoDB

[Zhang, Shuai]

Hi Alex,

Thank you very much for your prompt reply.

MongoDB harnesses GeoJSON to store Geospatial data and supports Geohash Index for spatial query as you mentioned in the link about the mongodb geospatial indexes. this project is trying to bring GDAL/OGR under the support of MongoDB. Technically, I think it's possible to write both vector and raster drivers, but I'm not sure whether I can handle both of them in a limited time.

As for the SOC idea page, I just cannot post the idea since login seems required. and i'm not sure how to get such access.

Thanks again,
Best,
shuai
________________________________________
From: Alex Mandel [tech_dev at wildintellect.com]
Sent: Monday, March 17, 2014 2:11 PM
To: Zhang, Shuai; soc at lists.osgeo.org
Subject: Re: [SoC] SoC Proposal on a OGR Driver for MongoDB

On 03/17/2014 11:23 AM, Zhang, Shuai wrote:
> Hi all,
>
> I wrote a proposal about a MongoDB Driver for GDAL/OGR, which has been submitted to the melange. I regard GSoC as a great opportunity to get connected with the Open Source world. Well, since it's my first time to do this, I'm really not sure whether it's proper. Here are some info.
>
> OGR Driver for MongoDB
> MongoDB, a document database that provides high performance, high availability, and easy scalability, can be a good platform for storing extremely large spatial datasets, to support high performance geo-computation and real-time spatial analysis in a large scale.This project aims at developing a OGR Driver for MongoDB to help applications or softwares based on GDAL, such as QGIS, Mapserver, Geoserver, Mapnik, ArcGIS and so on, read & write the spatial data in it, and thus enable the Open Source GIS Ecosystem powered by the advanced NoSQL database.
>
> A FEW WORDS ABOUT ME
> I'm a PhD candidate in my second year, with high performance geo-computation as my main research fields. This project is actually part of my research undergoing. when i saw the OSGeo GSoC flyer a week ago, i thought why not cooperate with the Open Source community and contribute my research. thereby, i wrote the proposal.
>
> I have spent a lot of time learning GDAL and building parallel geo-algorithms on it, so i think i'm quite familiar with GDAL/OGR and know how it works in the code level. I also devoted lots of my time in playing with MongoDB since we need a faster solution to support our high performance computing applications running on a large scale Linux cluster. the thought making GDAL/OGR communicate with MongoDB culster was just wonderful, simply because it prevents us from re-invent the wheel and get lots of GIS tools reusable.
>
> I'm looking forward to your advises.
> thank you for your time and consideration.
>
> My Regards,
> shuai
>
>

I assume you've started discussions about this topic on the gdal-dev
mailing list and added your idea to their SOC idea page? Please discuss
your idea there, as someone from GDAL/OGR would need to be the mentor
for the project.

Also how does this relate to the current implementation of geospatial
data storage in MongoDB? Writing a GDAL/OGR driver assumes that the
format already exists. Any reason it can't be both vector and raster?
I found http://docs.mongodb.org/manual/applications/geospatial-indexes/

These links also seems relevant:
http://www.paolocorti.net/2009/12/06/using-mongodb-to-store-geographic-data/
http://lists.osgeo.org/pipermail/gdal-dev/2013-February/035406.html

Thanks,
Alex