[GRASS-user] Processing Time for Vector Import w/OGR

GRASS GRASS user
1

   I'm importing the water bodies theme using v.in.ogr. There are 89,173
nodes and 99,129 primitives and boundaries. The process cleans polygons,
breaks polygons, and removes duplicates very quickly. Then it seems to take
forever (more than 40 minutes in one attempt) while breaking boundaries.

   This is on a moderately fast AMD-powered machine. Top shows two v.in.ogr
processes (one running 98 minutes, the other 12 minutes) consuming about 98%
of the CPU and about 23% of 2G RAM each process.

   Is this normal?

Rich

2

On Mon, 16 Nov 2009, Rich Shepard wrote:

I'm importing the water bodies theme using v.in.ogr. There are 89,173
nodes and 99,129 primitives and boundaries. The process cleans polygons,
breaks polygons, and removes duplicates very quickly. Then it seems to take
forever (more than 40 minutes in one attempt) while breaking boundaries.

   Update: I killed both processes and started anew. The v.in.ogr process to
break boundaries has been running for 10 minutes. Similar themes
(streams/rivers, hydrological units) completed much faster.

Rich

3

Rich Shepard wrote:

I'm importing the water bodies theme using v.in.ogr. There are 89,173
nodes and 99,129 primitives and boundaries. The process cleans polygons,
breaks polygons, and removes duplicates very quickly. Then it seems to take
forever (more than 40 minutes in one attempt) while breaking boundaries.

This is on a moderately fast AMD-powered machine. Top shows two v.in.ogr
processes (one running 98 minutes, the other 12 minutes) consuming about 98%
of the CPU and about 23% of 2G RAM each process.

Is this normal?

Unfortunately yes, better in grass7, see ticket #426
https://trac.osgeo.org/grass/ticket/426

Markus M

4

On Mon, 16 Nov 2009, Markus Metz wrote:

Unfortunately yes, better in grass7, see ticket #426
https://trac.osgeo.org/grass/ticket/426

Markus,

   OK.

   Should I be running 6.5 or 7 rather than the 6.4.0svn from last week?

Many thanks,

Rich

5

2009/11/16 Rich Shepard <rshepard@appl-ecosys.com>:

Should I be running 6.5 or 7 rather than the 6.4.0svn from last week?

In then best case the both, 6.5 for daily work, 7.x as experimental to
test new great features :wink:

Martin

--
Martin Landa <landa.martin gmail.com> * http://gama.fsv.cvut.cz/~landa

6

On Mon, 16 Nov 2009, Martin Landa wrote:

In then best case the both, 6.5 for daily work, 7.x as experimental to
test new great features :wink:

Martin,

   Thanks for the input. So 6.5 is sufficiently stable for production work
while 6.4 is still at RC5? Hmmm-m-m. That's interesting. :slight_smile:

   Perhaps this evening I'll build 6.5 and try it. Right now the CPU is
consumed with breaking boundaries on the standing water body theme for all
of Oregon ... I didn't realize there were so many lakes and reserviors!

Rich

7

Hi,

2009/11/16 Rich Shepard <rshepard@appl-ecosys.com>:

Thanks for the input. So 6.5 is sufficiently stable for production work
while 6.4 is still at RC5? Hmmm-m-m. That's interesting. :slight_smile:

There is not so big difference between 6.4 and 6.5. GRASS 6.5 just
brings some new features (GEOS support, etc.)

Martin

--
Martin Landa <landa.martin gmail.com> * http://gama.fsv.cvut.cz/~landa

8

On Mon, 16 Nov 2009, Martin Landa wrote:

There is not so big difference between 6.4 and 6.5. GRASS 6.5 just brings
some new features (GEOS support, etc.)

Martin,

   That makes good sense and explains what the web site means regarding back
ports of v.7 to v.6.

Thanks,

Rich

9

Rich Shepard wrote:

On Mon, 16 Nov 2009, Markus Metz wrote:

Unfortunately yes, better in grass7, see ticket #426
https://trac.osgeo.org/grass/ticket/426

Markus,

OK.

Should I be running 6.5 or 7 rather than the 6.4.0svn from last week?

The changes to v.in.ogr as in ticket #426 apply only to grass7. v.in.ogr in grass65 should take as long as in grass64, not much difference there AFAICT.

Markus M

10

On Tue, 17 Nov 2009, Markus Metz wrote:

The changes to v.in.ogr as in ticket #426 apply only to grass7. v.in.ogr
in grass65 should take as long as in grass64, not much difference there
AFAICT.

Markus,

   OK. That's what I thought after looking at the ticket. That theme -- water
bodies -- took 3 hours to finish breaking boundaries. I guess that when I do
the soils today I can expect similar times.

Thanks,

Rich