[GRASS-user] Best format for exporting raster data

[Rainer M Krug]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 01/09/10 19:45, Glynn Clements wrote:
> 
> Rainer M Krug wrote:
> 
>> It would be nice, if grass would be able to deal with on-the-fly
>> decompression - not from a .tar.gz file, but from gz compressed files.
> 
> GRASS rasters are already compressed by default, either using RLE or
> zlib compression (OTOH, the null bitmap isn't compressed; that will
> cease to be an issue if we embed nulls into the raster data).

That is good to know - as mentioned in an earlier email, I will then
leave as it is. I actually compared the size of the compresses grass
folder with the non-compressed one, and the difference did not justify
the effort.

> 
> But GRASS doesn't generally read data from files per se, but from
> either the GRASS "database" or from GDAL (and the former might
> eventually go away if we can get "native" GRASS support into GDAL).

I don't understand that statement - I am the database is ultimately a
file, so grass is requesting the data from the database which then reads
it from the mapset which consists of files. And gdal is used when data
is read which is not in the mapset - right?

> 
> The main issue with on-the-fly decompression using general-purpose
> formats is that rasters aren't guaranteed to be read sequentially,
> while compression algorithms require sequential access.

Makes sense - thanks for the clarification.

> 
> Similarly, while there exist filesystems which can mount archives, tar
> files (and especially compressed tar files) are a poor choice, as they
> are designed for sequential access. ZIP/RAR are more suited to such
> tasks.

but they can actually be mounted as was pointed out by Markus Neteler
(http://www.linux.com/archive/feature/132196).

> 
> Ultimately, I don't think that this situation is common enough to be
> worth doing anything about. If you want to access archived data, you
> just unpack the archives first (or use an archive format which can be
> mounted).

I'll folow the following approach: I'll leave the data uncompressed,
until the analysis are completed (I have a new 2TB hdd now) and will
compress them into tar.gz files as soon as the project is finished. If I
need them again, I will use archivemount to access the data.

Thanks everybody for your help,

Rainer

> 


- -- 
Rainer M. Krug, PhD (Conservation Ecology, SUN), MSc (Conservation
Biology, UCT), Dipl. Phys. (Germany)

Centre of Excellence for Invasion Biology
Natural Sciences Building
Office Suite 2039
Stellenbosch University
Main Campus, Merriman Avenue
Stellenbosch
South Africa

Tel:        +33 - (0)9 53 10 27 44
Cell:       +27 - (0)8 39 47 90 42
Fax (SA):   +27 - (0)8 65 16 27 82
Fax (D) :   +49 - (0)3 21 21 25 22 44
Fax (FR):   +33 - (0)9 58 10 27 44
email:      Rainer at krugs.de

Skype:      RMkrug
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkx/XyEACgkQoYgNqgF2egqrBwCggpE9+rv+KlfDBU3goF9lK0JK
GYUAn0nq2oam3Y2yG2tbRiZoQDfbLXou
=+nHe
-----END PGP SIGNATURE-----