Tuning query with sdo_join in it

Question

Hello all. I'm new to the forum, but I hope someone can help me tuning a spatial query. I have no DBA experience or education, but I'm writing SQL queries for a few years now, first in Oracle 9i (with Locator), now in Oracle 10g (10.2) and Oracle XE.
I’ve posted this in the Spatial forum as I suspect it’s mostly a spatial issue.
I have a table like:
CREATE TABLE CYCLORAMA (
  DATASETID   VARCHAR2(9 BYTE),
  IMAGEID     VARCHAR2(9 BYTE),
  OPNAMEDATUM CHAR(20 BYTE),
  GEOMETRIE   MDSYS.SDO_GEOMETRY
)with only a standard spatial index on GEOMETRIE, and containing about 180000 records. IMAGEID is unique, but not defined as such. Geometry is SDO_POINT_TYPE for all records, like: MDSYS.SDO_GEOMETRY(2001,90112,MDSYS.SDO_POINT_TYPE(86039.572,439541.158,'null'),'null','null'). USER_SDO_GEOM_METADATA is filled with values used for all my geometries: SDO_DIM_ARRAY(SDO_DIM_ELEMENT('X',0,300000,,001), SDO_DIM_ELEMENT('Y',300000,600000,,001)) and SRID = 90112.
What I need to do is to find (and ultimately delete) all records that have another record with a different value for DATASETID, a higher value for OPNAMEDATUM and fall within 3 meters distance. What I came up with so far is:
select /*+ all_rows */ s.imageid
, sdo_geom.sdo_distance(s.geometrie,t.geometrie,0.1) dist
from cyclorama s, cyclorama t
, table(sdo_join('CYCLORAMA','GEOMETRIE','CYCLORAMA','GEOMETRIE','distance=3')) x
-- create a crosstable with records that may be within 3 meters distance
where s.rowid = x.rowid1 and x.rowid2 = t.rowid and not x.rowid1 = x.rowid2
-- filter 'duplicate matches' sdo_join generates for a self join, and select:
and s.datasetid != t.datasetid                             -- if in different dataset
and s.opnamedatum < t.opnamedatum                          -- from a later date
and sdo_geom.sdo_distance(s.geometrie,t.geometrie,0.1) < 3 -- within 3 meters
;This is surprisingly fast (7 seconds for first 50 records), but still takes about 7 minutes (Oracle XE on a 768MB, 2.7GHz VM) for the whole query. This is acceptable, but:
* Is this solution the best approach?
* Can this be done faster or more efficient? Clever indexes, hints, smarter query, ...?
* Is it worth optimizing the SDO_DIM_ARRAY for this table, e.g. narrowing the range of coordinate values)?
* How to delete the found records most efficiently?
* Can this be done using an anti-join, to select the records to keep?
Thank you for reading this, and any answers you may have,
J-----.
Jeroen Muris

Answer 1

Jack,
Thank you for your response. It helped me to clean up the query. All the changes did not give a much better explain plan - at least not to my inexperienced eyes - but the total execution time for the query is now reduced to under two minutes. The query as it is now:
select /*+ ordered all_rows */ x.rowid1
from table(sdo_join('CYCLORAMA','GEOMETRIE','CYCLORAMA','GEOMETRIE','distance=3 mask=ANYINTERACT')) x
, cyclorama s, cyclorama t
where not x.rowid1 = x.rowid2
and s.rowid = x.rowid1 and x.rowid2 = t.rowid
and s.datasetid != t.datasetid
and s.opnamedatum < t.opnamedatum;Because the docs state that mask=FILTER is the default, I added an explicit mask=ANYINTERACT to the sdo_join parameters when I removed the sdo_distance from the query. Still, the query returns 205035 records with the sdo_distance, and 205125 without. But this may be the result of the extra 0.001 from sdo_dim. I did not investigate as the 3 meter is not crucial.
I believe I already had a mechanism in place to reduce the number of self-joins with "not x.rowid1 = x.rowid2" and "s.opnamedatum < t.opnamedatum". I can not guarantee that for all records s.opnamedatum < t.opnamedatum equals x.rowid1 < x.rowid2.
Based on the [url http://download.oracle.com/docs/cd/B19306_01/appdev.102/b14255/sdo_operat.htm#BGEDJIBF]docs, I finally added an 'ordered' hint and reshuffled the tables in the from clause.
I'm happy with performance now, and creating a new table with the records to keep should not be a problem. Still I'm curious about the following:
<li>Is it worth optimizing the SDO_DIM_ARRAY for the original table, e.g. narrowing the range of coordinate values?
<li>How can sdo_join best be used for an anti-join, to select the records to keep?
Thank you,
J-----.