Steve<br><br>I was pursuing the performance issue on the pmapper mailing list too. It transpires it is related to browser performance (from a JS aspect) for the web apps query/redraw processes -> IE7 - awful (>8mins), IE8 and FF pretty good - around 20secs for returning/highlighting 5000 features (and building results table) and Chrome about 13 secs, that is without even creating a spatial index. Thanks for the info and I think I may even not have to go down the db route (which keeps life simple!) just advise IE6/7 users to switch browsers (not the first webGIS app I know that does this....!)<br>
<br>Thanks<br>Chris.<br><br><div class="gmail_quote">On 25 August 2010 18:08, Lime, Steve D (DNR) <span dir="ltr"><<a href="mailto:Steve.Lime@state.mn.us">Steve.Lime@state.mn.us</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div link="blue" vlink="purple" lang="EN-US">
<div>
<p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);">Like with everything it depends… Selection of
features via a bounding box shouldn’t be much, or really any, slower than
drawing since it’s basically the same mechanism for both. Attribute
queries can be quite slow with shapefiles since MapServer doesn’t do any
attribute indexing. That’s where the RDMS really adds value. There
shouldn’t be much of a difference with a bbox query, and in fact, a local
shapefile will probably be faster, even on larger datasets as long as you’ve
build an spatial index file.</span></p>
<p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);"> </span></p>
<p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);">Anyway, it’s what you do with results afterwards that
matters and that’s probably specific to your application. I’ve
used query files to persist selected features over many map draws successfully
on modest (50,000 feature) sized shapefiles.</span></p>
<p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);"> </span></p>
<p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);">Steve</span></p>
<p class="MsoNormal"><span style="font-size: 11pt; color: rgb(31, 73, 125);"> </span></p>
<div style="border-width: 1pt medium medium; border-style: solid none none; border-color: rgb(181, 196, 223) -moz-use-text-color -moz-use-text-color; padding: 3pt 0in 0in;">
<p class="MsoNormal"><b><span style="font-size: 10pt;">From:</span></b><span style="font-size: 10pt;"> Chris Jackson
[mailto:<a href="mailto:webturtles@gmail.com" target="_blank">webturtles@gmail.com</a>] <br>
<b>Sent:</b> Tuesday, August 24, 2010 10:34 AM<br>
<b>To:</b> Lime, Steve D (DNR)<br>
<b>Cc:</b> <a href="mailto:mapserver-users@lists.osgeo.org" target="_blank">mapserver-users@lists.osgeo.org</a><br>
<b>Subject:</b> Re: [mapserver-users] Feature select performance</span></p>
</div><div><div></div><div class="h5">
<p class="MsoNormal"> </p>
<p class="MsoNormal" style="margin-bottom: 12pt;">Steve<br>
<br>
The selection of layer data is done by a bounding box, though querying by
parameter values is just as slow. The mapfile is directly calling up the
shapefile (I had just followed the P.Mapper examples of displaying/querying
shapefiles directly). So probably the answers to those questions lie
under the hood of P.Mapper, which I assume is using the PHP mapscript.dll to do
the work. Is the direct query to shapefile approach v.efficient for larger
datasets..?. Ignoring the app I am using what is a good approach to large
feature datasets from a MapServer perspective?<br>
<br>
Thanks<br>
Chris<br>
<br>
<br>
</p>
<div>
<p class="MsoNormal">On 24 August 2010 14:27, Lime, Steve D (DNR) <<a href="mailto:Steve.Lime@state.mn.us" target="_blank">Steve.Lime@state.mn.us</a>>
wrote:</p>
<p class="MsoNormal">How is the selection done? How is the highlighting done?
Those processes will really impact performance depending on how things are
implemented.<br>
<br>
Steve<br>
<br>
________________________________________<br>
From: <a href="mailto:mapserver-users-bounces@lists.osgeo.org" target="_blank">mapserver-users-bounces@lists.osgeo.org</a>
[<a href="mailto:mapserver-users-bounces@lists.osgeo.org" target="_blank">mapserver-users-bounces@lists.osgeo.org</a>]
On Behalf Of Chris Jackson [<a href="mailto:webturtles@gmail.com" target="_blank">webturtles@gmail.com</a>]<br>
Sent: Tuesday, August 24, 2010 3:19 AM<br>
To: <a href="mailto:mapserver-users@lists.osgeo.org" target="_blank">mapserver-users@lists.osgeo.org</a><br>
Subject: [mapserver-users] Feature select performance</p>
<div>
<div>
<p class="MsoNormal"><br>
Hi all<br>
<br>
I have rigged up a webGIS app (P.Mapper framework) where one of the underlying
datasets is a model grid (many thousand cells) in a shapefile. The data
displays nice and quick when just viewing, but if I allow the user to select
say 5000 cells it is very (extremely) slow to redraw with the highlighted
cells. Doing the same task in a very old ArcGIS server web app was much
quicker (admittedly the data was in SDE/MS SQL). So my question is what
is the best way to improve performance - would just transferring the shapefile
to to say a MS SQL 2005 db with the MSSQLSPATIAL add-on be a good step or is
there are better way to present the data in the Map file?<br>
Many thanks<br>
Chris</p>
</div>
</div>
</div>
<p class="MsoNormal"> </p>
</div></div></div>
</div>
</blockquote></div><br>