Re: [sqlite] Bugreport - slowdown in sqlite after the ANALYZE statement

On Wed, 10 Dec 2014 08:49:21 +0100 Eduardo Morras wrote: > > Why ORDER BY on INSERT? Does it work better? I would expect the > > unnecessary sort to be pure overhead. > > If you insert in correct index order, the index update phase is > faster because it don't need

Re: [sqlite] Bugreport - slowdown in sqlite after the ANALYZE statement

Thank you very much for the explanation and tips, they are appreciated. Dne 9.12.2014 v 14:30 Richard Hipp napsal(a): > Answered by adding a comment at > https://bugzilla.redhat.com/show_bug.cgi?id=1161844 > > On Tue, Dec 9, 2014 at 6:06 AM, Jan Staněk wrote: > > Hi, > some

Re: [sqlite] Bugreport - slowdown in sqlite after the ANALYZE statement

On Tue, 9 Dec 2014 10:38:34 -0500 "James K. Lowden" wrote: > On Tue, 09 Dec 2014 12:06:20 +0100 > Jan Stan?k wrote: > > > INSERT INTO CoreCache (ModelID, ItemID) > > SELECT > ... > > ORDER BY Year > > Why ORDER BY on INSERT? Does it work

Re: [sqlite] Bugreport - slowdown in sqlite after the ANALYZE statement

On Tue, 09 Dec 2014 12:06:20 +0100 Jan Stan?k wrote: > INSERT INTO CoreCache (ModelID, ItemID) > SELECT ... > ORDER BY Year Why ORDER BY on INSERT? Does it work better? I would expect the unnecessary sort to be pure overhead. --jkl

Re: [sqlite] Bugreport - slowdown in sqlite after the ANALYZE statement

Answered by adding a comment at https://bugzilla.redhat.com/show_bug.cgi?id=1161844 On Tue, Dec 9, 2014 at 6:06 AM, Jan Staněk wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Hi, > some of the banshee users noticed a huge slowdown in its operation > after

[sqlite] Bugreport - slowdown in sqlite after the ANALYZE statement

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, some of the banshee users noticed a huge slowdown in its operation after upgrading to version 3.8.7 from 3.8.6. Here is the related log : [4 Debug 13:24:27.263] Executed in 12819ms DELETE FROM CoreCache WHERE ModelID = 9;