You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I noticed a significant performance difference when running a script using the conda windows build compared to the binaries available on sqlite.org.
The last statement in the script below takes about double the time: 31 seconds compared to 18 seconds.
Could there be a difference in the compilation parameters that explains this difference?
-- Create table for bboxes
CREATE TABLE bboxes (
id INTEGER PRIMARY KEY AUTOINCREMENT NOT NULL,
minx FLOAT,
maxx FLOAT,
miny FLOAT,
maxy FLOAT
);
-- Fill the table with 1 million bboxes
INSERT INTO bboxes(minx, maxx, miny, maxy)
WITH RECURSIVE
cnt(x) AS (
SELECT 0
UNION ALL
SELECT x+10 FROM cnt
LIMIT 1000
)
SELECT minx, maxx, miny, maxy
FROM (SELECT x AS minx, x+10 AS maxx FROM cnt) x,
(SELECT x AS miny, x+10 AS maxy FROM cnt) y;
-- Create and fill up rtree.
-- Takes +- 31 secs using conda sqlite.exe, +- 18 secs with version downloaded from sqlite.org
CREATE VIRTUAL TABLE bboxes_rtree USING rtree(id, minx, maxx, miny, maxy);
INSERT INTO bboxes_rtree
SELECT id, minx, maxx, miny, maxy FROM bboxes;
Comment:
I noticed a significant performance difference when running a script using the conda windows build compared to the binaries available on sqlite.org.
The last statement in the script below takes about double the time: 31 seconds compared to 18 seconds.
Could there be a difference in the compilation parameters that explains this difference?
Reference to OSGeo/gdal#7614
The text was updated successfully, but these errors were encountered: