On Mon, Apr 24, 2006 at 08:54:40PM -0400, Paul Malcher wrote:
> On 4/24/06, Jim C. Nasby <[EMAIL PROTECTED]> wrote:
> >
> > Keep in mind, there's *WAY* more Windows users than Linux users; that
> > doesn't mean Windows is a better OS.
>
>
> Indeed
's a few
applications that are difficult to get good performance out of with
PostgreSQL. Website session tables that are constantly updated are one
example.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
something). I don't
believe the same can be said of MySQL.
If you do decide to go with MySQL, at least do yourself a favor and use
InnoDB and turn all the strict mode stuff on.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.com
need
to do some testing to see how much it helped you. (And indeed, how much
different schemes helped you).
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
identify some
common groupings, and partition based on that. Even if you can't, you
could at least get a win on single-person queries.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
o me like the best bet is to put the 'resources' into a database
and let it handle the locking...
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
te the old data and refresh it (or simply overwrite it). In doing so
> your other applications would then see a new time stamp within the accepted
> threshold range and so could now trust that data again.
>
> Wayne
>
>
> "Jim C. Nasby" <[EMAIL PROTECTED]&
sue a COMMIT. Note that some databases won't like you
leaving that transaction open a real long time, so it depends on what
you're doing if this will work. I also don't know if SQLite cares about
such things.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasiv
and
running, not provide good performance.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
saction cannot modify or lock rows changed
by other transactions after the serializable transaction began.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
te when
> finished, if needed delete "processed" records.
FWIW, the performance of that would be pretty bad in most MVCC
databases, because you can't do an update 'in place' (Ok, Oracle can,
but they still have to write both undo and redo log info, so it's
effectively
ups that allready part of the same path.
> In this way i get an endless recursion (1).
>
> GROUP-->SUB-->SUB-->SUB-->PRODUCT
> | | |
> +-<-1-+ +--->SUB-->PRODUCT
>
> Is there a way to check all parent (sub)groubs inside a trigger?
&
nk "Feb 31" is a valid date,
maybe that's OK...
Their information about using a nested set model seems accurate, though.
Another option is to use 'ltree'. There used to be an implimentation for
PostgreSQL, but it looks like it's been removed.
http://developer.postgresq
On Fri, Feb 10, 2006 at 04:44:30PM +0100, Fanda Vacek wrote:
> You can choose what to use. An Elephant or a Dolphin :))
Or you could use both! :P
http://commandprompt.com/images/mammoth_versus_dolphin_500.jpg
are
> trying to put information that should be in a joined table into columns in
> the main table.
Agreed. What is it you're trying to do?
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
ng in
the table. So when you need the info most you can't get it.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
sys.com> Voice: 503-667-4517 Fax:
> 503-667-8863
>
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
t;
> >Fanda
> >
> >On Tue, 07 Feb 2006 22:35:09 +0100, John Stanton <[EMAIL PROTECTED]>
> >wrote:
> >
> >>Jim C. Nasby wrote:
> >>
> >>>On Mon, Feb 06, 2006 at 05:30:39PM +0100, Laurent Goussard wrote:
> >>>
> >>
of application.
Sounds pretty different. Flightaware tracks all flights currently in the
air that are under the control of the air traffic system (basically,
commercial flights and anyone who's filed an IFR flight plan). I
actually know the folks behind the site.
--
Jim C. Nasby, Sr. Engineer
. An
added benefit of a version field is it's a trivial way to detect
potential conflicts; if you have the edit code pass back the version it
was editing, you can verify that that version is what's still in the
database.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Per
ou want to test both sync and async options you can turn fsync
off in PostgreSQL by setting fsync to false. If you do that I'd also
turn full_page_writes off as well, since there's no point in writing the
extra info.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Softw
On Tue, Feb 07, 2006 at 08:07:03PM +0100, Nemanja Corlija wrote:
> On 2/7/06, Jim C. Nasby <[EMAIL PROTECTED]> wrote:
> > AFAIK MySQL ships with a few different config files, so presumably
> > choosing the appropriate one would be equivalent to what I provided for
> &g
On Tue, Feb 07, 2006 at 11:51:22AM -0500, [EMAIL PROTECTED] wrote:
> "Jim C. Nasby" <[EMAIL PROTECTED]> wrote:
> > On Tue, Feb 07, 2006 at 09:54:22AM -0600, Jim C. Nasby wrote:
> > > Hrm, that's rather odd. What does top show when it's running through
&g
On Tue, Feb 07, 2006 at 08:11:02PM +0100, Nemanja Corlija wrote:
> On 2/7/06, Jim C. Nasby <[EMAIL PROTECTED]> wrote:
> > I see theh tcl now... is TCL piping into psql, or are there a set of raw
> > files you could post or send me?
> TCL generates files and then I
On Tue, Feb 07, 2006 at 10:04:43AM -0500, Clay Dowling wrote:
>
> Jim C. Nasby said:
>
> > Well, that certainly won't help things... at a minimum, on your machine,
> > you should change the following:
> > shared_buffers=1
> > effective_cache_size=10
On Tue, Feb 07, 2006 at 09:54:22AM -0600, Jim C. Nasby wrote:
> Hrm, that's rather odd. What does top show when it's running through
> psql? Are the test scripts available for download? I'll try this on my
> machine as well...
I see theh tcl now... is TCL piping into psq
resumably
choosing the appropriate one would be equivalent to what I provided for
PostgreSQL.
BTW, has anyone looked at adding SQLite support to any of the DBT
benchmarks? http://sourceforge.net/projects/osdldbt
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Softw
On Tue, Feb 07, 2006 at 10:08:23AM +0100, Nemanja Corlija wrote:
> On 2/7/06, Jim C. Nasby <[EMAIL PROTECTED]> wrote:
> > On Tue, Feb 07, 2006 at 07:31:50AM +0100, Nemanja Corlija wrote:
> > > > Well, that certainly won't help things... at a minimum, on your machin
for bringing the database up on a
machine with very minimal hardware. It's the equivalent to using MySQL's
minimum configuration file.
It certainly doesn't seem unreasonable to tweak a handful of parameters
for each database. I wouldn't even consider this to be tu
On Tue, Feb 07, 2006 at 07:06:26AM +0100, Nemanja Corlija wrote:
> On 2/7/06, Jim C. Nasby <[EMAIL PROTECTED]> wrote:
> > For test 8 on PostgreSQL, what's EXPLAIN ANALYZE for one of those show?
> test=# EXPLAIN ANALYZE SELECT count(*), avg(b) FROM
of those show?
What changes have you made to the default postgresql.conf?
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
.)
FWIW, I believe SQLite's syntax is closer to PostgreSQL's than it is to
MySQL, so it might me easier to migrate that direction...
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
On Tue, Jan 31, 2006 at 10:05:47AM -0700, Dennis Cote wrote:
> [EMAIL PROTECTED] wrote:
> CREATE TABLE x(i INTEGER PRIMARY KEY AUTOINCREMENT CHECK(i < (1<<32)));
I suspect you'll see better performance if you hard-code the value
instead of doing a bit-shift every time you inse
on)
> >is going to be slow.
>
> Heh, that should obviously read '6 million rows' . Sorry, its early.
Unless your rows were exceptionally wide, that shouldn't be difficult to
handle at all.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive S
ing up with missing log files, I would certainly hope
it would just stop dead in it's tracks rather than corrupt anything.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasiv
On Mon, Jan 09, 2006 at 09:08:10PM -0800, Dan Kennedy wrote:
>
>
> --- "Jim C. Nasby" <[EMAIL PROTECTED]> wrote:
>
> > On Mon, Jan 09, 2006 at 06:47:04PM +0100, Eduardo wrote:
> > > of transactions per second. But because each transaction requires
On Mon, Jan 09, 2006 at 06:47:04PM +0100, Eduardo wrote:
> of transactions per second. But because each transaction requires at
> least two revolutions of the disk platter, SQLite is limited to about
Why does a transaction commit require two seperate writes?
--
Jim C. Nasby, Sr. Engin
tries in order?
You can either just use a simple sequence and then delete where sequence
mod = 0 (hrm... does sqlite support
functional indexes?), or just add a timestamp to the table.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://perva
IQUE constraint violation occurs, the pre-existing rows that
are causing the constraint violation are removed prior to inserting or
updating the current row.
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http
; end;
Wouldn't that be a lot more efficient with some kind of EXISTS test
rather than a count(*)?
--
Jim C. Nasby, Sr. Engineering Consultant [EMAIL PROTECTED]
Pervasive Software http://pervasive.comwork: 512-231-6117
vcard: http://jim.nasby.net/pervasive.vcf cell: 512-569-9461
7;);
would fail, but
INSERT INTO students(age) VALUES('18');
should work. Of course, you're ultimately just making the database do
more work; you should really just insert the int as an int and be done
with it...
INSERT INTO students(age) VALUES(18);
--
Jim C. Nasby, Sr.
WHERE id =
(SELECT id
FROM tbl1
42 matches
Mail list logo