Lincoln Yeoh <[EMAIL PROTECTED]> writes:
> Why is SHMMAX so low on some O/Ses?
Historical artifact, I think: the SysV IPC code was developed on
machines that were tiny by current standards. Unfortunately, vendors
haven't stopped to review their kernel parameters and scale them up
appropriately.
On Fri, 22 Jun 2001, Alex Pilosov wrote:
> On Fri, 22 Jun 2001, Thomas T. Thai wrote:
>
> > > Huh? That *is* the generic, fully-SQL-standard way.
> >
> > the project i'm doing this for is opensource and i need it to work in both
> > postgresql and mysql. the above statement wouldn't work in my
some quick (and I might add disappointing) benchmark results...
must tune..
.
will be performing -and publishing results- of more comprehensive and
rigorous suite on:
- 2x700MHz cpu
- 4GB RAM
- 6x35GB 10k rpm RAID 5 system
- linux 2.4.4
- postgres 7.1.2-2
I've perf-tuned many an oracle db -
"Thomas T. Thai" <[EMAIL PROTECTED]> writes:
> On Fri, 22 Jun 2001, Tom Lane wrote:
>
> > "Thomas T. Thai" <[EMAIL PROTECTED]> writes:
> > >> Use a union query:
> > >>
> > >> select rec_id, path, '' as link, name from cat_cat
> > >> UNION
> > >> select rec_id, path, link, name from cat_alias
>
On Fri, 22 Jun 2001, Tom Lane wrote:
> "Thomas T. Thai" <[EMAIL PROTECTED]> writes:
> >> Use a union query:
> >>
> >> select rec_id, path, '' as link, name from cat_cat
> >> UNION
> >> select rec_id, path, link, name from cat_alias
>
> > there is no way to do this in a generic DBI way? i need f
Peter Eisentraut <[EMAIL PROTECTED]> writes:
> Tom Lane writes:
>> This does remind me that I'd been thinking of suggesting that we
>> raise the default -B to something more reasonable, maybe 1000 or so
>> (yielding an 8-meg-plus shared memory area).
> On Modern(tm) systems, 8 MB is just as arbit
Bruce Momjian <[EMAIL PROTECTED]> writes:
>> We could offer a --with switch to manually choose the default, too.
> Good idea, yes. Not sure if we need a --with switch because they can
> just edit the postgresql.conf or postgresql.conf.sample file.
Well, we have a --with switch for DEF_MAXBACKEN
Tried with 2048 also, it complete took away the strange steep after 7:
D is now 2048
1.A: 36B: 32C: 35D: 31
2.A: 69B: 53C: 38D: 38
3.A: 97B: 79C: 40D: 40
4.A: 131B: 98C: 48D: 43
5.A: 163B: 124C: 52D: 49
6.A: 210
Bruce Momjian <[EMAIL PROTECTED]> writes:
>> Hmm. Anyone like the idea of a platform-specific default established
>> by configure? We could set it in the template file on platforms where
>> the default SHMMAX is too small to allow 1000 buffers.
> Template file seems like a good idea for platfor
Bruce Momjian <[EMAIL PROTECTED]> writes:
> Strange that even at 1024 performance still drops off at 7. Seems it
> may be more than buffer thrashing.
Yeah, if anything the knee in the curve seems to be worse at 1024
buffers. Curious. Deserves more investigation, perhaps.
This does remind me t
I'll take advantage of the fact the Pg is "open source" and that Bruce comments on the
story, so as not to be flamed about this:
http://news.cnet.com/news/0-1003-200-6352301.html?tag=prntfr
Q: What are 1M lawyers at the bottom of the sea?
A: A good start
cheers,
thalis
ps any lawyers in the
Bruce Momjian <[EMAIL PROTECTED]> writes:
>> This does remind me that I'd been thinking of suggesting that we
>> raise the default -B to something more reasonable, maybe 1000 or so
>> (yielding an 8-meg-plus shared memory area).
> BSD/OS has a 4MB max but we document how to increase it by recompi
"Thomas T. Thai" <[EMAIL PROTECTED]> writes:
>> Use a union query:
>>
>> select rec_id, path, '' as link, name from cat_cat
>> UNION
>> select rec_id, path, link, name from cat_alias
> there is no way to do this in a generic DBI way? i need for this to work
> across diff kind of DBs.
Huh? That
On Fri, 22 Jun 2001, Bruce Momjian wrote:
> > I would really like is to have the "User's lounge" link changed to "User
> > Documentation". If I haven't been to the postgresql site in a few months,
> > when I look at the "User's lounge" link, I always think, "Is that a chat
> > room or something?"
Hi again,
I have been following this list for approx. 6600 messages (i think
that is 2 months or so.. .) - an yes.. Like the Cardassian Order I
keep everything :)
A couple of time, ppl has been talking about speed..
What about a couple of benchmarks from your systems, so that us
newbies (or
On Fri, 22 Jun 2001, Edwin Grubbs wrote:
> Unless you have some kind of development environment that requires it, you
> should avoid using inheritance. Besides the fact that I recently asked on
> the list on how to get indexes to work when selecting from all the
> inherited tables at once and sti
On Fri, 22 Jun 2001 [EMAIL PROTECTED] wrote:
>
>
> Use a union query:
>
> select rec_id, path, '' as link, name from cat_cat
> UNION
> select rec_id, path, link, name from cat_alias
there is no way to do this in a generic DBI way? i need for this to work
across diff kind of DBs.
>
> Notice
"Thalis A. Kalfigopoulos" wrote on Fri, 22 Jun 2001 16:34:56 EDT
>On Fri, 22 Jun 2001, Alex Pilosov wrote:
>
> [SNIP]
>
>Had the same problem with a disk and boosted its performance with hdparm (4->3
>5Mb/s). I don't know if this was the reason I started getting the message "--
>MARK --" in my sy
man syslogd
-m interval
The syslogd logs a mark timestamp regularly default 20 minutes.
No virus *sigh* :)
Daniel Åkerud
- Original Message -
From: "Thalis A. Kalfigopoulos" <[EMAIL PROTECTED]>
To: "Alex Pilosov" <[EMAIL PROTECTED]>
Cc: "Paul" <[EMAIL PROTECTED]>; <[EMAIL PROT
Holy ultra-violet-active macaronies :)
First I changed it to 256, then I changed it to 1024.
-B 128 is A
-B 256 is B
-B 1024 is C
New multiple-index performance data):
1.A: 36B: 32C: 35
2.A: 69B: 53C: 38
3.A: 97B: 79C: 40
4.A: 131B: 98C: 48
5.
On Fri, 22 Jun 2001, Alex Pilosov wrote:
> Whoa! Your drive is only capable of 3M/s (640M/213 sec) transfer! Your
> data set is 500K*200 bytes=100M. No surprise your queries take 30 seconds
> to run, since its exactly how much it'd take to transfer 100M of data from
> your drive.
>
> Please read
Hi,
I administer an quite important PGsql v. 7.0.3 installation... We
would very much like to upgrade, BUT we cannot afford to be offline
for a very long period .. half an hour is acceptable.
I have tried to install the new one under a new location and have
done an init-db.. both are running
22 matches
Mail list logo
