what about the differences between the FULL and PARTIAL modes of pragma
synchronous ?
--
View this message in context:
http://www.nabble.com/how-can-i-speed-up-inserts---tf3186848.html#a8860860
Sent from the SQLite mailing list archive at Nabble.com.
On 2/8/07, Dennis Cote <[EMAIL PROTECTED]> wrote:
DragonK wrote:
>
>
> Well, using transactions would be a little difficult, because I have a
> library which does something like logging to a database. Using
> transaction
> on a single insert would be useless in my opinion, and on the other
>
On 2/7/07, ohadp <[EMAIL PROTECTED]> wrote:
luckily this isn't mission critical failure, i'll take the risk that one
out
of ten thousand users experiences database corruption.
i can only batch the inserts into maybe 4-5 a time, don't know how much of
an improvement that will be...
If
ohadp wrote:
when you say much less do you mean one in 1,000 or 1 in 1,000,000 ?
Gerry Snyder-3 wrote:
If you really can't get more than 4-5 in a batch, it's your call whether
the time saving is worth the risk. BTW, the problem rate should be much
less than one in ten thousand.
Sorry. I
when you say much less do you mean one in 1,000 or 1 in 1,000,000 ?
Gerry Snyder-3 wrote:
>
> If you really can't get more than 4-5 in a batch, it's your call whether
> the time saving is worth the risk. BTW, the problem rate should be much
> less than one in ten thousand.
>
> Gerry
>
>
DragonK wrote:
Well, using transactions would be a little difficult, because I have a
library which does something like logging to a database. Using
transaction
on a single insert would be useless in my opinion, and on the other
hand I
can't use transactions on the all inserts, since
On 2/7/07, John Stanton <[EMAIL PROTECTED]> wrote:
DragonK wrote:
> On 2/7/07, Ken <[EMAIL PROTECTED]> wrote:
>
>>
>> Try without the pragma and wrap the inserts with a begin transaction
>> and a
>> commit...
>>
>> The performance will be almost as good as with the pragma, with the
>> added
Hello DragonK,
Essentially, when you have 1000's of uses banging on your software,
many with poorly maintained/infected machines things that seldom
happen on well maintained PC's will happen all the time. Some of my
users have daily Windows crashes and assume that it's a normal
experience for
DragonK wrote:
On 2/7/07, Ken <[EMAIL PROTECTED]> wrote:
Try without the pragma and wrap the inserts with a begin transaction
and a
commit...
The performance will be almost as good as with the pragma, with the
added benefit of consistent data and no corruption in the event of a
crash
Wed, Feb 07, 2007 at 07:36:10PM +0200: DragonK wrote:
> On 2/7/07, Teg <[EMAIL PROTECTED]> wrote:
> >
> >Hello ohadp,
> >
> >Well, my experience is, particularly when it's users using it and not
> >a personal project, that corruption happens fairly frequently when you
> >use this pragma. That's
On 2/7/07, Ken <[EMAIL PROTECTED]> wrote:
Try without the pragma and wrap the inserts with a begin transaction and a
commit...
The performance will be almost as good as with the pragma, with the
added benefit of consistent data and no corruption in the event of a crash
or power failure.
I
Try without the pragma and wrap the inserts with a begin transaction and a
commit...
The performance will be almost as good as with the pragma, with the added
benefit of consistent data and no corruption in the event of a crash or power
failure.
DragonK <[EMAIL PROTECTED]> wrote:
On 2/7/07, Teg <[EMAIL PROTECTED]> wrote:
Hello ohadp,
Well, my experience is, particularly when it's users using it and not
a personal project, that corruption happens fairly frequently when you
use this pragma. That's why I don't use it any more in my production
code.
Transactions are far
Dennis Cote schrieb:
ohadp wrote:
holy smokes that pragma just made my several hundred inserts take half a
second compared to 30 seconds earlier.
thanks!
Yes, but now your database can be corrupted by a power failure or OS crash.
Wrapping several hundred inserts in a transaction will be
ohadp wrote:
luckily this isn't mission critical failure, i'll take the risk that one out
of ten thousand users experiences database corruption.
Only you can judge how important your data are.
i can only batch the inserts into maybe 4-5 a time, don't know how much of
an improvement that will
Hello ohadp,
Well, my experience is, particularly when it's users using it and not
a personal project, that corruption happens fairly frequently when you
use this pragma. That's why I don't use it any more in my production
code.
Transactions are far safer and fast too.
C
Wednesday, February 7,
Try it. You might find that it executes in a similar time and it will
still be ACID and safe.
ohadp wrote:
luckily this isn't mission critical failure, i'll take the risk that one out
of ten thousand users experiences database corruption.
i can only batch the inserts into maybe 4-5 a time,
luckily this isn't mission critical failure, i'll take the risk that one out
of ten thousand users experiences database corruption.
i can only batch the inserts into maybe 4-5 a time, don't know how much of
an improvement that will be...
Dennis Cote wrote:
>
> ohadp wrote:
>> holy smokes that
holy smokes that pragma just made my several hundred inserts take half a
second compared to 30 seconds earlier.
thanks!
DragonK wrote:
>
> On 2/7/07, Ohad Eder-Pressman <[EMAIL PROTECTED]> wrote:
>>
>> i've got sqlite compiled with visual-studio, with all the default
>> options,
>> didn't
On 2/7/07, Ohad Eder-Pressman <[EMAIL PROTECTED]> wrote:
i've got sqlite compiled with visual-studio, with all the default options,
didn't touch anything.
inserts are just a bit too slow for me, is there a good way to speed this
up
using some flags ?
any other flags that it would be smart to
i've got sqlite compiled with visual-studio, with all the default options,
didn't touch anything.
inserts are just a bit too slow for me, is there a good way to speed this up
using some flags ?
any other flags that it would be smart to turn on for some better
performance ?
thanks
21 matches
Mail list logo
