> ... See included png file.
What kind of machine was this run on?
- Thomas
Tatsuo Ishii <[EMAIL PROTECTED]> writes:
> I agree with Tom. I did some benchmarking tests using pgbench for a
> computer magazine in Japan. I got a almost equal or better result for
> 7.1 than 7.0.3 if commit_delay=0. See included png file.
Interesting curves. One thing you might like to know i
> "Schmidt, Peter" <[EMAIL PROTECTED]> writes:
> > So, is it OK to use commit_delay=0?
>
> Certainly. In fact, I think that's about to become the default ;-)
I agree with Tom. I did some benchmarking tests using pgbench for a
computer magazine in Japan. I got a almost equal or better result for
I wrote:
> Thus, our past arguments about whether a few microseconds of delay
> before commit are a good idea seem moot; we do not have any portable way
> of implementing that, and a ten millisecond delay for commit is clearly
> Not Good.
I've now finished running a spectrum of pgbench scenarios,
> >
> > BTW, is 7.1 going to be a bit slower than 7.0? Or just Beta 5? Just
> > curious. Don't mind waiting for 7.2 for the speed-up if necessary.
>
> It is possible that it will be ... the question is whether the slow down
> is unbearable or not, as to whether we'll let it hold things up or not
> > No, but I haven't looked at it. I am now much more concerned with the
> > delay, and am wondering if I should start thinking about trying my idea
> > of looking for near-committers and post the patch to the list to see if
> > anyone likes it for 7.1 final. Vadim will not be back in enough ti
The Hermit Hacker <[EMAIL PROTECTED]> writes:
> No way to group the writes to you can keep the most recent one open?
> Don't see an easy way, do you?
>>
>> No, but I haven't looked at it. I am now much more concerned with the
>> delay,
I concur. The blind write business is not important enough
Tatsuo Ishii <[EMAIL PROTECTED]> writes:
> Probably we should create the regression database
> using template0?
Seems like a good idea.
regards, tom lane
> Other than that, I have nothing to hold up a beta5. Anyone else?
>
> regards, tom lane
I see a small problem with the regression test. If PL/pgSQL has been
already to template1, the regression scripts will fail because
createlang fails. Probably we should create the regr
On Fri, 16 Feb 2001, Bruce Momjian wrote:
>
> > Bruce Momjian <[EMAIL PROTECTED]> writes:
> > >> But I'm not going to touch it for 7.1...
> >
> > > No way to group the writes to you can keep the most recent one open?
> >
> > Don't see an easy way, do you?
> >
>
> No, but I haven't looked at it.
On Sat, 17 Feb 2001, Lincoln Yeoh wrote:
> At 04:17 PM 2/16/01 -0500, Tom Lane wrote:
> >
> >Vadim says (and I agree) that we really ought to implement a new
> >lightweight lock manager that would fall between spinlocks and regular
> >locks in terms of overhead and functionality. But it's not re
"Schmidt, Peter" <[EMAIL PROTECTED]> writes:
> So, is it OK to use commit_delay=0?
Certainly. In fact, I think that's about to become the default ;-)
I have now experimented with several different platforms --- HPUX,
FreeBSD, and two considerably different strains of Linux --- and I find
that t
> BTW, is 7.1 going to be a bit slower than 7.0? Or just Beta 5? Just
> curious. Don't mind waiting for 7.2 for the speed-up if necessary.
>
We expect 7.1 to be faster than 7.0.X. We may have a small problem that
we may have to address. Not sure yet.
--
Bruce Momjian
> Bruce Momjian <[EMAIL PROTECTED]> writes:
> >> But I'm not going to touch it for 7.1...
>
> > No way to group the writes to you can keep the most recent one open?
>
> Don't see an easy way, do you?
>
No, but I haven't looked at it. I am now much more concerned with the
delay, and am wonder
Bruce Momjian <[EMAIL PROTECTED]> writes:
>> But I'm not going to touch it for 7.1...
> No way to group the writes to you can keep the most recent one open?
Don't see an easy way, do you?
regards, tom lane
At 04:17 PM 2/16/01 -0500, Tom Lane wrote:
>
>Vadim says (and I agree) that we really ought to implement a new
>lightweight lock manager that would fall between spinlocks and regular
>locks in terms of overhead and functionality. But it's not reasonable
Will there be an arbitrary user locking fe
> After looking a little more, I think this is too big a change to risk
> making for beta. I was thinking it might be an easy change, but it's
> not; there's noplace to store the open-relation reference if we don't
> have a relcache entry. But we don't want to pay the price of opening a
> relcac
Bruce Momjian <[EMAIL PROTECTED]> writes:
>> I am thinking it'd be a good idea if blind write went through fd.c and
>> thus was able to re-use open file descriptors, just like normal writes.
>> This should improve the efficiency of dumping dirty buffers during
>> checkpoint by a noticeable amount.
> 3. Blind write is gratuitously inefficient: it does separate open,
> seek, write, close kernel calls for every request. This was the right
> thing in 7.0.*, because backends relatively seldom did blind writes and
> even less often needed to blindwrite multiple pages of a single relation
> in su
While poking at Peter Schmidt's comments about pgbench showing worse
performance than for 7.0 (using -F in both cases), I noticed that given
enough buffer space, FileWrite never seemed to get called at all. A
little bit of sleuthing revealed the following:
1. Under WAL, we don't write dirty buff
I am GO. SET DIAGNOSTICS is my only open item left.
> The Hermit Hacker <[EMAIL PROTECTED]> writes:
> > things appear to have quieted off nicely ... so would like to put out a
> > Beta5 for testing ...
>
> > Tom, I saw/read your proposal about the JOIN syntax, but haven't seen any
> > commit o
At 12:15 16/02/01 -0500, Bruce Momjian wrote:
>
>Can someone comment on this? Seems GET DIAGNOSTICS is wrong from
>Peter's reading of SQL92, and mine too.
>
He's quite correct.
Philip Warner| __---_
Al
> -Original Message-
> From: Tom Lane [mailto:[EMAIL PROTECTED]]
>
> Peter Eisentraut <[EMAIL PROTECTED]> writes:
> > The GNU C library now offers a %a (and %A) format that prints floating
> > point numbers in a semi-internal form that is meant to be portable. (I
> > image this was done b
> (Using ColId instead of datetime + IDENT gives reduce/reduce conflicts
> that I don't want to mess with now.)
> The date_part implementation is prepared for unknown field selectors, so
> this should be all safe. Comments?
Works for me. Since extract required explicit reserved words, I had just
The Hermit Hacker <[EMAIL PROTECTED]> writes:
> things appear to have quieted off nicely ... so would like to put out a
> Beta5 for testing ...
> Tom, I saw/read your proposal about the JOIN syntax, but haven't seen any
> commit on it yet, nor any arguments against the changes ... so just
> wonde
Perhaps someone can explain what's wrong with this. Excuse the mess,
it was cut out of a much larger function but reliably creates the
error on my Postgres 7.1 beta 4 machine.
Compile C function, restart postgres (for the heck of it), create a
new db (I used 'stuff), import sql. The insert it r
On Fri, 16 Feb 2001, Martin A. Marques wrote:
> Hi
> I'd like to know if someone is working on the ALTER TABLE specs of Postgres.
> I think that on this field there is much to do, especially comparing with
> other DB servers, which let you change lots of more things that Postgres
> doesn't.
>
The age() function is documented as "Calculate time difference while
retaining year/month fields", but it doesn't seem to do anything different
from a plain date subtraction with a few time zone problems added in:
select age(date '1999-05-17', date '1957-06-13');
age
---
Online -- Goa writes:
> I have a Win32 application that uses ODBC to connect to the Postgres
>(ver
> 6.5) on RedHat Linux dbase. I have been able to connect and perform a lot
> of SQL statements, etc. However I am unable to perform the backup of the
> dbase. I need to backup up the
things appear to have quieted off nicely ... so would like to put out a
Beta5 for testing ...
Tom, I saw/read your proposal about the JOIN syntax, but haven't seen any
commit on it yet, nor any arguments against the changes ... so just
wondering where those stand right now?
Marc G. Fournier
Hi,
They're a little dated, but Great Bridge funded some benchmarks last
summer putting Postgres 7.0 against Unnamed Proprietary Database 1
(version 8i) and Unnamed Proprietary Database 2 (version 7.0, available
for NT platform only). See
http://www.greatbridge.com/about/press.php?content_id
ISTM that it is mighty confusing that extract() and date_part() don't
accept the same set of "field" arguments.
-> SELECT EXTRACT(decade FROM TIMESTAMP '2001-02-16 20:38:40');
ERROR: parser: parse error at or near "decade"
=> SELECT EXTRACT("decade" FROM TIMESTAMP '2001-02-16 20:38:40');
ERROR:
Hello!!
I need demostrate that PostgreSQL is a great RDBMS for my undergraduate
project, because this, Does somebody has a bechmark (or similar
document) between Postgres and others DB (commercial DB's, principally)?
Thanks in advance!!
Dear friends,
I have been searching the mailing lists for a couple of days now hoping to
find a solution to my problem. Well I hope I find a solution here.
The problem is such:
I have a Win32 application that uses ODBC to connect to the Postgres
(ver
6.5) on RedHat Linu
Hello,
There is a bug in the binary distribution for S.u.S.E. 7.0; in the script
"/etc/rc.d/postgres", in the "start" clause.
The -D option of the postmaster daemon is used to declare where is the data
directory.
You do it like this:
postgres -D$datadir
but you must do it like this:
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> The GNU C library now offers a %a (and %A) format that prints floating
> point numbers in a semi-internal form that is meant to be portable. (I
> image this was done because of C99, but I'm speculating.) It might be
> useful to offer this to preserv
Hi,
I realy get into problems witht his one.
I've made an Irix nss library which connects to postgresql.
But somehow the backend doesn;t get into active status.
The blocking PQconnectdb halts until timeout (if i ignore the
errormessage, the results return right after the timeout has expired).
Hi there...
I would like to inquire of possible support for running PostgreSQL on a
Linux Cluster. How would i implement and configure PostgreSQL as a
distributed database i.e. replicated on several servers?
I am anxious to hear from you guys.
--
Manny C. Ca
Hello world,
I saw that PostgreSQL doesn't works with
a MOSIX cluster because shared memory.
Well, some day MOSIX will be there, I'll
be waiting for a shared memory support!
The question is: does anybody know another
cluster software that supports shared
memory?
Thanx,
--
---
I'm trying to figure out the SPI and need a bit of help, because the
docs do not mention (that I can find) some parts of the interface.
- SPI_exec(char * sql, int count): this function seems to execute the
query. Is the `count' parameter the maximum number of tuples to
return? Does count=0
Hello,
I'm writing a name server switch (nss) module for Irix/PostgreSQL, but i
have some trouble to get it to work.
When i use the blocking functions of PostgreSQL, then there isn't any
response until the nss daemons times out. But right after that i get the
results (which is too late in this s
On Fri, Feb 16, 2001 at 01:02:24PM -0500, Tom Lane wrote:
> Adam Haberlach <[EMAIL PROTECTED]> writes:
> > I had postgres start blocking all it's UPDATEs on a production
> > database today, when an engineer added the following two tables,
> > among other things. We've had to restore from back
Adam Haberlach <[EMAIL PROTECTED]> writes:
> I had postgres start blocking all it's UPDATEs on a production
> database today, when an engineer added the following two tables,
> among other things. We've had to restore from backup, and the
> interesting thing is that when we re-add these tab
Peter T Mount <[EMAIL PROTECTED]> writes:
> Just a quick question, but how much of SQL92 Entry Level does 7.1
> support, and what parts haven't we got (yet)?
I don't think anyone's made a careful list --- making one is on my
personal to-do list for the near future, but not yet at the top.
Schema
I had postgres start blocking all it's UPDATEs on a production
database today, when an engineer added the following two tables,
among other things. We've had to restore from backup, and the
interesting thing is that when we re-add these tables, things
break again.
Version: Postg
Can someone comment on this? Seems GET DIAGNOSTICS is wrong from
Peter's reading of SQL92, and mine too.
> Bruce Momjian writes:
>
> > > Bruce Momjian writes:
> > >
> > > > New PL/pgSQL GET DIAGNOSTICS statement for SPI value access (Jan)
> > >
> > > If someone can show me an example of how i
Hiroshi Inoue writes:
> Is there a way to change the precision of floating
> point representation from clients ?
Not currently, but I image it couldn't be too hard to introduce a
parameter that changes the format string used by float*out to something
else.
The GNU C library now offers a %a (and
Just a quick question, but how much of SQL92 Entry Level does 7.1 support, and
what parts haven't we got (yet)?
I need to know for a couple of internal bits in the JDBC driver...
Peter
--
Peter Mount [EMAIL PROTECTED]
PostgreSQL JDBC Driver: http://www.retep.org.uk/postgres/
RetepPDF PDF libr
> [EMAIL PROTECTED] writes:
>
> > What showed up was the "Error index_formtuple: data takes 21268 bytes: too
> > big". If anyone has any ideas on this, please share them.
>
> It means your data is too big to fit into an index.
Good case in point. Here is a typical email. Here is a difficult/
No. In parser/gram.y I see:
| MATCH PARTIAL
{
elog(ERROR, "FOREIGN KEY/MATCH PARTIAL not yet implemented");
$$ = "PARTIAL";
}
> MATCH PARTIAL isn't in 7.1. Is it?
>
> Mike Mascari
> [EMAIL PROTECTED]
>
>
--
Bruce Momjian
MATCH PARTIAL isn't in 7.1. Is it?
Mike Mascari
[EMAIL PROTECTED]
> Tatsuo Ishii <[EMAIL PROTECTED]> writes:
> > Oh, I see. So the question still remains: can a MULTIBYTE-aware backend
> > ever use a sort order different from strcmp() order? (That is, not as
> > a result of LOCALE, but just because of the non-SQL-ASCII encoding.)
> >>
> >> According to the co
52 matches
Mail list logo
