[Mapserver-users] raster data formats

Ed McNierney ed at topozone.com
Tue Jun 1 20:07:28 EDT 2004


Bart -

There is no "ideal" solution, because there is no "ideal" problem.  It's
a bit like asking what the most ideal automobile is - a lot depends on
what you need to do with it.  The "most ideal raster format" depends
upon many factors:

	In what format is the data available?
	How difficult is it to convert the data to another format?
	Must you preserve the original data exactly?
	How much disk space can you afford?
	What user load and response time requirements do you have?

CPU and memory are lesser factors, because it's usually relatively cheap
to get a "lot" of them, whereas multiple terabytes of disk space still
costs quite a bit of money - relatively speaking, of course.

The third question is important - YOU need to know whether you can
consider lossy compression or not.  If you need to preserve the original
data pixel values exactly, then you simply cross all the lossy
compression formats off the list.

There's nothing out there to "convince people" - you need to understand
your problem, understand the various alternatives, and make the best
choice.

	- Ed

Ed McNierney
President and Chief Mapmaker
TopoZone.com
-----Original Message-----
From: mapserver-users-admin at lists.gis.umn.edu
[mailto:mapserver-users-admin at lists.gis.umn.edu] On Behalf Of Bart van
den Eijnden
Sent: Tuesday, June 01, 2004 4:34 PM
To: mapserver-users at lists.gis.umn.edu
Subject: [Mapserver-users] raster data formats

Hi list,

did somebody do a (recent) research into what would be the most ideal
raster format for Mapserver (or for webmapping in general)?

Aspects which come to mind:

-time to produce the map
-CPU usage
-disk space
-memory consumption

I would especially be interested in comparing compressed Geotiff,
uncompressed Geotiff, ECW and JPEG2000.

I know Ed McNierney has made several strong pleas for uncompressed
Geotiff although depending on the situation, but are there any hard
figures out there to convince people? Ofcourse every research will
depend on the rasters used, but just wondering if there are any figures
out there.

Thanks in advance.

Best regards,
Bart
_______________________________________________
Mapserver-users mailing list
Mapserver-users at lists.gis.umn.edu
http://lists.gis.umn.edu/mailman/listinfo/mapserver-users






More information about the mapserver-users mailing list