[TCMUG] Best processes for visualizing large datasets?

[Jennifer Strahan]

I've been out of the loop for the past few years, staying home with my now
2 year old and am  trying to get back up to speed on best practices for
visualizing and interacting with large datasets (eg. Twin Cities parcels,
Foursquare data, etc.).  I'm working with a local consulting group that
does a lot of site selection analysis and similar projects.

I've been asked to give some recommendations on server requirements and
technology approach and wonder if you all might have some suggestions.

Our requirements are:
1.  Ability to connect with dynamic data sources
2.  Generate vector tiles
3.  Create dynamic charts/graphs - using tools such as d3 (or whatever is
preferred these days)
4.  3D map rendering

I started playing with MapBox and looks like we can do a lot with it, but
it doesn't offer everything we need and seems like it could get expensive.
I also wonder if I should play with Carto.  In the past, I've worked with
PostGIS, MapServer, Tilestache and had access to a server that was managed
by an admin person.  I'm wondering if I should recommend that they get a
server with the open source stack installed, or if we should rely on
services.  If a server is recommended, are there packages that make it easy
to install and manage software updates?

Would love to hear your suggestions!

Thanks,
Jennifer
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/tcmug/attachments/20170507/b3fc7eef/attachment.html>