mirror of
https://git.postgresql.org/git/postgresql.git
synced 2024-12-21 08:29:39 +08:00
b90f8f20f0
to the existing X-direction tests. An rtree class now includes 4 actual 2-D tests, 4 1-D X-direction tests, and 4 1-D Y-direction tests. This involved adding four new Y-direction test operators for each of box and polygon; I followed the PostGIS project's lead as to the names of these operators. NON BACKWARDS COMPATIBLE CHANGE: the poly_overleft (&<) and poly_overright (&>) operators now have semantics comparable to box_overleft and box_overright. This is necessary to make r-tree indexes work correctly on polygons. Also, I changed circle_left and circle_right to agree with box_left and box_right --- formerly they allowed the boundaries to touch. This isn't actually essential given the lack of any r-tree opclass for circles, but it seems best to sync all the definitions while we are at it. |
||
|---|---|---|
| .. | ||
| bench | ||
| data | ||
| expected | ||
| sql | ||
| Makefile | ||
| README.rtree_gist | ||
| rtree_gist.c | ||
| rtree_gist.sql.in | ||
This is a R-Tree implementation using GiST.
Code (for PG95) are taken from http://s2k-ftp.cs.berkeley.edu:8000/gist/pggist/
and changed according to new version of GiST (7.1 and above)
All work was done by Teodor Sigaev (teodor@stack.net) and Oleg Bartunov
(oleg@sai.msu.su). See http://www.sai.msu.su/~megera/postgres/gist
for additional information.
CHANGES:
Oct 10 MSD 2001
1. Implemented new linear algorithm for picksplit
ref. ( 'New Linear Node Splitting Algorithm for R-tree',
C.H.Ang and T.C.Tan )
Tue May 29 17:04:16 MSD 2001
1. Small fixes in polygon code
Thanks to Dave Blasby <dblasby@refractions.net>
Mon May 28 19:42:14 MSD 2001
1. Full implementation of R-tree using GiST - gist_box_ops,gist_poly_ops
2. gist_poly_ops is lossy
3. NULLs support
4. works with multi-key GiST
NOTICE:
This version will only work with PostgreSQL version 7.1 and above
because of changes in the function call interface.
INSTALLATION:
gmake
gmake install
-- load functions
psql <database> < rtree_gist.sql
REGRESSION TEST:
gmake installcheck
EXAMPLE USAGE:
create table boxtmp (b box);
-- create index
create index bix on boxtmp using gist (b gist_box_ops);
-- query
select * from boxtmp where b && '(1000,1000,0,0)'::box;
BENCHMARKS:
subdirectory bench contains benchmark suite.
Prerequisities: perl, DBI, DBD:Pg, Time::HiRes
cd ./bench
1. createdb TEST
2. psql TEST < ../rtree_gist.sql
3. ./create_test.pl | psql TEST
-- change $NUM - number of rows in test dataset
4. ./bench.pl - perl script to benchmark queries.
Run script without arguments to see available options.
a)test without GiST index, using built-in R-Tree
./bench.pl -d TEST
b)test R-Tree using GiST index
./bench.pl -d TEST -g
RESULTS:
1. One interesting thing is that insertion time for built-in R-Tree is
about 8 times more than ones for GiST implementation of R-Tree !!!
2. Postmaster requires much more memory for built-in R-Tree
3. Search time depends on dataset. In our case we got:
+------------+-----------+--------------+
|Number boxes|R-tree, sec|R-tree using |
| | | GiST, sec |
+------------+-----------+--------------+
| 10| 0.002| 0.002|
+------------+-----------+--------------+
| 100| 0.002| 0.002|
+------------+-----------+--------------+
| 1000| 0.002| 0.002|
+------------+-----------+--------------+
| 10000| 0.015| 0.025|
+------------+-----------+--------------+
| 20000| 0.029| 0.048|
+------------+-----------+--------------+
| 40000| 0.055| 0.092|
+------------+-----------+--------------+
| 80000| 0.113| 0.178|
+------------+-----------+--------------+
| 160000| 0.338| 0.337|
+------------+-----------+--------------+
| 320000| 0.674| 0.673|
+------------+-----------+--------------+