[mapserver-users] SDE performance and error

[Stephen Lime]

You read just the parts you need. It's been a while but nodes and leafs have
a fixed size so you can seek your way around the index by computing the 
the location of things. Can't remember if Carl optimized that or not but it seems
pretty efficient.

Stephen Lime
Internet Applications Analyst

Minnesota DNR
500 Lafayette Road
St. Paul, MN 55155
651-297-2937

>>> Frank Warmerdam <warmerdam at pobox.com> 07/12/01 04:32PM >>>
> The result after shptree using default parameters was great.  I can now get
> rid of the 2 files and use vdotrd.
> 
> If I get the time to benchmark against SDE I will let you know.  I'm
> installing Oracle 8.1.7 and SDE 8.1 as we speak on a SUN E4500 (8 400mhz
> CPUs, 8 mb ECache each CPU and 4 GB RAM). This should be a nice test.

Lowel,

Sweet machine!

> I assume you wrote this index?  It's based on R-tree or Quad-tree type
> algorithms as opposed to ESRI's indexing scheme?

The algorithm is a simple quad tree.  The only clever aspect is that I
general a moderate overlap between quads to ensure that small objects that
overlap a quad boundary don't necessarily get promoted too far up the quad
tree.  

I haven't checked how Steve has done the quadtrees on disk, but in an 
ideal world only portions of the index file that are required would be
read.  Otherwise you end up consuming at least something like 4 or 8 bytes
of memory for every shape just to hold the index.  However, this would be 
OK for medium sized files. 

Best regards,

---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com 
light and sound - activate the windows | http://pobox.com/~warmerdam 
and watch the world go round - Rush    | Geospatial Programmer for Rent