More newbie questions... Is there a need for PostGIS, or is PostgreSQL
enough? I found that "GN doesn't really store spatial data or do queries
against it, except for the extents of records. It's a metadata plant."
Perhaps PostGIS does not at least disturb, but is it necessary?
Arto
Hi,
2010/3/31 Arto Vuorela <arto.k.vuorela@anonymised.com>
More newbie questions... Is there a need for PostGIS, or is PostgreSQL
enough? I found that "GN doesn't really store spatial data or do queries
against it, except for the extents of records. It's a metadata plant."
Perhaps PostGIS does not at least disturb, but is it necessary?
I just configured a geonetwork trunk instance against a PostGresql database
without postgis support (no lwpostgis.sql loaded, nor plpgsql installed on
the db). That seems to work like a charm. So I'd guess that postgis is
unnecessary ;
PS : Why not adding by default the PostGreSQL support into the config.xml
file ? Since the jdbc driver for postgres is already into the svn trunk.
--- config.xml (revision 5905)
+++ config.xml (working copy)
@@ -115,8 +115,25 @@
<poolSize>10</poolSize>
</config>
</resource>
+
+ <!-- - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - -->
+ <!-- postgres -->
+ <!-- - - - - - - - - - - - - - - - - - - - - - - - - - - -
- - - -->
+ <resource enabled="false">
+ <name>main-db</name>
+ <provider>jeeves.resources.dbms.DbmsPool</provider>
+ <config>
+ <user>www-data</user>
+ <password>www-data</password>
+ <driver>org.postgresql.Driver</driver>
+
<url>jdbc:postgresql://localhost:5432/geonetwork</url>
+ <poolSize>10</poolSize>
+ <reconnectTime>3600</reconnectTime>
+ </config>
+ </resource>
</resources>
+
Cheers,
--
Pierre Mauduit
Camptocamp France SAS
Savoie Technolac, BP 352
73377 Le Bourget du Lac Cedex
Tel : + 33 (0)4 79 44 44 92
http://www.camptocamp.com
pierre.mauduit@anonymised.com
No, definitively not.
My understanding is that spatial information stored as such exists only in the shapefile extents index, as of 2.4.
The advent of 3.x and a more sophisticated persistence subsystem may or may not change that. I am not qualified to speak to that. I would be curious to hear the current thinking from the GN core team on that topic. Will spatial indexing move into the DB layer?
---
A. Soroka
Digital Research and Scholarship R & D
the University of Virginia Library
On Mar 31, 2010, at 6:22 AM, Arto Vuorela wrote:
More newbie questions... Is there a need for PostGIS, or is PostgreSQL
enough? I found that "GN doesn't really store spatial data or do queries
against it, except for the extents of records. It's a metadata plant."
Perhaps PostGIS does not at least disturb, but is it necessary?Arto
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-users
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork
PostGis extension is needed if you wont store geographic data
(vectorial) in a PostGres DBMS . Metadata is not geographic data, but
only data about other data.
So...... Postgres is enough.
2010/3/31 <ajs6f@anonymised.com>:
No, definitively not.
My understanding is that spatial information stored as such exists only in the shapefile extents index, as of 2.4.
The advent of 3.x and a more sophisticated persistence subsystem may or may not change that. I am not qualified to speak to that. I would be curious to hear the current thinking from the GN core team on that topic. Will spatial indexing move into the DB layer?
---
A. Soroka
Digital Research and Scholarship R & D
the University of Virginia LibraryOn Mar 31, 2010, at 6:22 AM, Arto Vuorela wrote:
More newbie questions... Is there a need for PostGIS, or is PostgreSQL
enough? I found that "GN doesn't really store spatial data or do queries
against it, except for the extents of records. It's a metadata plant."
Perhaps PostGIS does not at least disturb, but is it necessary?Arto
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-users
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-users
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork
--
Ing. Francesco D'Amore
blog: www.marinajonica.org
linkedin: http://www.linkedin.com/in/francescodamore
Consiglio Nazionale delle Ricerche - Istituto di Inquinamento Atmosferico (IIA)
Ph. D. Student at University of Calabria - DEIS
Hi,
At present (2.4.x and 2.5), the spatial index is only used by CSW which needs to get answers to queries like here is a polygon, give me all records that have an extent that overlaps/touches/disjoint/etc. If you do a search from the user interface with an area of interest/bounding box drawn on the minimap then the spatial index is not used, instead range queries are done against bbox extent numbers that are in the Lucene index.
There is an enhancement for 2.5 (which might also be applied to 2.4.3) that has just been passed by the PSC - see PerformanceEnhancements2 – GeoNetwork opensource Developer website - that does move spatial indexing of extents* to the DB layer - but at present only if you set up GeoNetwork with PostGIS (which has been added to gast as a db option) - Oracle Spatial should come later.
The main reason this has been done is it that some operations need to access the shapes in the shapefile through an ID attribute (eg. during delete/update). Without an index (indexed access to shapefile attributes in geotools is not yet integrated with the shapefile datastore), this can be pretty slow when the number of metadata records in the catalog starts to get above 20,000 (very slow by 40,000 on my machine). So in short, if you expect to have a catalog of that size (think about all those records you might harvest!) then PostGIS is definitely your friend.
Cheers,
Simon
* extent can be one or more bounding boxes and/or polygons in the enhancement.
________________________________________
From: ajs6f@anonymised.com [ajs6f@anonymised.com]
Sent: Thursday, 1 April 2010 1:16 AM
To: geonetwork-users Users
Subject: Re: [GeoNetwork-users] Need for PostGIS?
No, definitively not.
My understanding is that spatial information stored as such exists only in the shapefile extents index, as of 2.4.
The advent of 3.x and a more sophisticated persistence subsystem may or may not change that. I am not qualified to speak to that. I would be curious to hear the current thinking from the GN core team on that topic. Will spatial indexing move into the DB layer?
---
A. Soroka
Digital Research and Scholarship R & D
the University of Virginia Library
On Mar 31, 2010, at 6:22 AM, Arto Vuorela wrote:
More newbie questions... Is there a need for PostGIS, or is PostgreSQL
enough? I found that "GN doesn't really store spatial data or do queries
against it, except for the extents of records. It's a metadata plant."
Perhaps PostGIS does not at least disturb, but is it necessary?Arto
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
Best Open Source Mac Front-Ends 2024
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
geonetwork-users List Signup and Options
GeoNetwork OpenSource is maintained at GeoNetwork - Geographic Metadata Catalog download | SourceForge.net
Thank you, Mr. Pigot, for this useful clarification.
May I ask further?
I understand at this time from:
http://trac.osgeo.org/geonetwork/wiki/Hibernate
that it is intended to collapse the Lucene index into a Hibernate-supported persistence layer. May we have any expectation of using Hibernate Spatial to collapse, in parallel, the spatial index in the same way in order to remove any need to support in particular PostGIS, Oracle Spatial, etc.?
Our site currently uses PostGIS to back GN as a matter of convenience (we use PostGIS as part of our backend to a massive GeoServer plant), but we contemplate a number of possible future directions for that data store. We would rather design towards a point of convergence (i.e. Hibernate Spatial) than a particular product.
And (with your patience): how will this interact with an eventual move towards ebRIM and an ESB architecture that might segregate a search engine entirely?
---
A. Soroka
Digital Research and Scholarship R & D
the University of Virginia Library
On Mar 31, 2010, at 7:00 PM, <Simon.Pigot@anonymised.com> wrote:
Hi,
At present (2.4.x and 2.5), the spatial index is only used by CSW which needs to get answers to queries like here is a polygon, give me all records that have an extent that overlaps/touches/disjoint/etc. If you do a search from the user interface with an area of interest/bounding box drawn on the minimap then the spatial index is not used, instead range queries are done against bbox extent numbers that are in the Lucene index.
There is an enhancement for 2.5 (which might also be applied to 2.4.3) that has just been passed by the PSC - see http://trac.osgeo.org/geonetwork/wiki/PerformanceEnhancements2 - that does move spatial indexing of extents* to the DB layer - but at present only if you set up GeoNetwork with PostGIS (which has been added to gast as a db option) - Oracle Spatial should come later.
The main reason this has been done is it that some operations need to access the shapes in the shapefile through an ID attribute (eg. during delete/update). Without an index (indexed access to shapefile attributes in geotools is not yet integrated with the shapefile datastore), this can be pretty slow when the number of metadata records in the catalog starts to get above 20,000 (very slow by 40,000 on my machine). So in short, if you expect to have a catalog of that size (think about all those records you might harvest!) then PostGIS is definitely your friend.
Cheers,
Simon* extent can be one or more bounding boxes and/or polygons in the enhancement.
________________________________________
From: ajs6f@anonymised.com [ajs6f@anonymised.com]
Sent: Thursday, 1 April 2010 1:16 AM
To: geonetwork-users Users
Subject: Re: [GeoNetwork-users] Need for PostGIS?No, definitively not.
My understanding is that spatial information stored as such exists only in the shapefile extents index, as of 2.4.
The advent of 3.x and a more sophisticated persistence subsystem may or may not change that. I am not qualified to speak to that. I would be curious to hear the current thinking from the GN core team on that topic. Will spatial indexing move into the DB layer?
---
A. Soroka
Digital Research and Scholarship R & D
the University of Virginia LibraryOn Mar 31, 2010, at 6:22 AM, Arto Vuorela wrote:
More newbie questions... Is there a need for PostGIS, or is PostgreSQL
enough? I found that "GN doesn't really store spatial data or do queries
against it, except for the extents of records. It's a metadata plant."
Perhaps PostGIS does not at least disturb, but is it necessary?Arto
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-users
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
GeoNetwork-users mailing list
GeoNetwork-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-users
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork