[sqlite] Query times vary between 0.2 s and 30 s for very

2015-03-24 Thread James K. Lowden
On Mon, 23 Mar 2015 06:43:39 + Hick Gunter wrote: > SQLite creates an ephemeral table for the IN list,giving O(log n) > performance for lookups. Thank you for pointing that out, Hick. Good to know! Is there some lower bound on either the size of the IN list or the number of rows in the t

[sqlite] What is wrong with this simple SQL?

2015-03-24 Thread James K. Lowden
On Sun, 22 Mar 2015 19:04:50 -0400 Stephen Chrzanowski wrote: > What I'm gathering from Igor and Ryan is that even though the subquery > looks to be a self contained entity, it actually isn't? Right, it actually isn't. All the names mentioned in the outer query are in scope for the inner one $

[sqlite] Very poor SQLite performance when using Win8.1 + Intel RAID1

2015-03-24 Thread R.Smith
On 2015-03-24 05:43 PM, Rob van der Stel wrote: > Hello, > > Currently I am investigating a SQLite performance problem that started to > occur when we switched from using Windows XP to Windows 8.1 for our > applications. > The following information is obtained to comparing two hardware identic

[sqlite] Problem: you can delete a virtual table with open statements and then SEGV

2015-03-24 Thread Richard Hipp
On 3/24/15, Roger Binns wrote: > On 03/23/2015 04:33 PM, Roger Binns wrote: >> SQLite really should prevent it from happening by disallowing the >> deletion of virtual tables with open statements. > > Thanks for fixing this. I can confirm that the latest SQLite dev code > now gives the expected d

[sqlite] Very poor SQLite performance when using Win8.1 + Intel RAID1

2015-03-24 Thread Keith Medcalf
Seems about right to me. If I do 1000 inserts on an SQLite database backed on spinning disk with a 32GB Flash Cache it takes 3 seconds. I would expect that (assuming your SSD are using a Connor interface) you would see at least twice that time used for RAID 1. Is your new Intel RAID driver t

[sqlite] Problem: you can delete a virtual table with open statements and then SEGV

2015-03-24 Thread Roger Binns
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 03/23/2015 04:33 PM, Roger Binns wrote: > SQLite really should prevent it from happening by disallowing the > deletion of virtual tables with open statements. Thanks for fixing this. I can confirm that the latest SQLite dev code now gives the expe

[sqlite] Very poor SQLite performance when using Win8.1 + Intel RAID1

2015-03-24 Thread Rob van der Stel
Hello, Currently I am investigating a SQLite performance problem that started to occur when we switched from using Windows XP to Windows 8.1 for our applications. The following information is obtained to comparing two hardware identical systems one running Win XP the other running Win 8.1. ***

[sqlite] Very poor SQLite performance when using Win8.1 + Intel RAID1

2015-03-24 Thread Simon Slavin
On 24 Mar 2015, at 3:43pm, Rob van der Stel wrote: > Samsung 128 GB SSD Are those Samsung 830s ? Those things are insanely fast. FlushFileBuffers() does all the hard work in disk access. Other calls may end up changing only cached information whereas FlushFileBuffers() has to go mess with

[sqlite] Very poor SQLite performance when using Win8.1 + Intel RAID1

2015-03-24 Thread Jeremy Nicoll - ml sqlite users
Rob van der Stel wrote: >Hello, > > Currently I am investigating a SQLite performance problem that started to > occur when we switched from using Windows XP to Windows 8.1 for our > applications. I don't know anything about SQLite, but faced with what you're seeing I think I'd want to be sure h

[sqlite] Query times vary between 0.2 s and 30 s for very

2015-03-24 Thread Mario M. Westphal
The temporary table is creates as CREATE TEMPORARY TABLE _tempNNN (oid INTEGER PRIMARY KEY) So the optimizer must know that the values are unique.