Re: [PERFORM] Moving postgresql.conf tunables into 2003...

2003-07-20 Thread Bruce Momjian
a directive I haven't used for a while. Or the comments are enough to let me know I don't need a directive, or that I need to go to the manual and read more. I appreciate that. Isn't that what we have now --- isn't postgresql.conf clear enough to jog people's memory. -- Bruce Momjian

Re: [PERFORM] PostgreSQL vs. MySQL

2003-07-20 Thread Bruce Momjian
. ---(end of broadcast)--- TIP 7: don't forget to increase your free space map settings -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road

Re: [PERFORM] PostgreSQL vs. MySQL

2003-07-20 Thread Bruce Momjian
as a good consultant I complied. I'm not going to touch if that was a smart instruction to give :-) Performance is probably 'good enough', and you can revisit it later when you have more time. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED

Re: [PERFORM] [pgsql-advocacy] About the default performance

2003-07-20 Thread Bruce Momjian
/FAQ.html -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end

Re: [PERFORM] Help disk-pages

2003-07-21 Thread Bruce Momjian
! - ---(end of broadcast)--- TIP 4: Don't 'kill -9' the postmaster -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your

Re: [PERFORM] Tunning FreeeBSD and PostgreSQL

2003-07-21 Thread Bruce Momjian
allocate page tables and prevent that address range from being used by other processes. Of course, if you have much less than 4 gigs of RAM in the machine, it probably isn't an issue. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359

Re: [PERFORM] PGAvd

2003-07-21 Thread Bruce Momjian
... -- -Josh Berkus Aglio Database Solutions San Francisco ---(end of broadcast)--- TIP 8: explain analyze is your friend -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359

Re: [PERFORM] Dual Xeon + HW RAID question

2003-07-22 Thread Bruce Momjian
doesn't seem like much. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073

Re: [PERFORM] Tunning FreeeBSD and PostgreSQL

2003-07-22 Thread Bruce Momjian
Vivek Khera wrote: BM == Bruce Momjian [EMAIL PROTECTED] writes: not anything pre-allocated (from my understanding). These settings allow for up to 100,000 shared buffers (I currently only use 30,000 buffers) BM I think the only downside to making them too big is that you allocate

Re: [PERFORM] Dual Xeon + HW RAID question

2003-07-22 Thread Bruce Momjian
about going from 1 to 1.4 or 2 to 2.8, which doesn't seem like much. ---(end of broadcast)--- TIP 8: explain analyze is your friend -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610

Re: [PERFORM] Dual Xeon + HW RAID question

2003-07-22 Thread Bruce Momjian
Jord Tanner wrote: On Tue, 2003-07-22 at 10:39, Bruce Momjian wrote: But CPU affinity isn't realated to hyperthreading, as far as I know. CPU affinity tries to keep processes on the same cpu in case there is still valuable info in the cpu cache. It is true that CPU affinity

Re: [PERFORM] Perfomance Tuning

2003-08-14 Thread Bruce Momjian
. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast

Re: [PERFORM] Perfomance Tuning

2003-08-14 Thread Bruce Momjian
others. --- Reece Hart wrote: On Mon, 2003-08-11 at 15:16, Bruce Momjian wrote: That _would_ work if ext2 was a reliable file system --- it is not. Bruce- I'd like to know your evidence for this. I'm not refuting

Re: [PERFORM] On Linux Filesystems

2003-08-14 Thread Bruce Momjian
. --- Christopher Browne wrote: Bruce Momjian commented: Uh, the ext2 developers say it isn't 100% reliable ... I mentioned it while I was visiting Red Hat, and they didn't refute it. 1. Nobody has gone through any formal proofs, and there are few

Re: [PERFORM] Filesystems WAS: Perfomance Tuning

2003-08-14 Thread Bruce Momjian
your free space map settings -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073

Re: [PERFORM] Perfomance Tuning

2003-08-14 Thread Bruce Momjian
--- it is not. This is the problem of Linux file systems --- they have unreliable, and journalled, with nothing in between, except using a journalling file system and having it only journal metadata. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359

Re: [PERFORM] On Linux Filesystems

2003-08-14 Thread Bruce Momjian
. --- Christopher Browne wrote: Bruce Momjian commented: Uh, the ext2 developers say it isn't 100% reliable ... I mentioned it while I was visiting Red Hat, and they didn't refute it. 1. Nobody has gone through any formal proofs

Re: [PERFORM] The results of my PostgreSQL/filesystem performance tests

2003-08-30 Thread Bruce Momjian
system atomicity anymore anyway. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073

Re: [PERFORM] Hardware recommendations to scale to silly load

2003-08-30 Thread Bruce Momjian
. If you have a battery-backed controller, you should be able to do the same. (You will not need to turn fsync off --- fsync will just be fast because of the disk drive RAM). Am I missing something? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED

Re: [PERFORM] Tests

2003-08-30 Thread Bruce Momjian
of broadcast)--- TIP 5: Have you checked our extensive FAQ? http://www.postgresql.org/docs/faqs/FAQ.html -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life

Re: [PERFORM] FreeBSD page size

2003-09-03 Thread Bruce Momjian
Marc G. Fournier wrote: On Wed, 3 Sep 2003, Bruce Momjian wrote: Vivek Khera wrote: the restore complained often about checkpoints occurring every few seconds: Sep 2 11:57:14 d02 postgres[49721]: [5-1] LOG: checkpoints are occurring too frequently (15 seconds apart) Sep

Re: [PERFORM] FreeBSD page size

2003-09-04 Thread Bruce Momjian
number is from Tatsuo testing a few years ago. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania

Re: [PERFORM] FreeBSD page size

2003-09-04 Thread Bruce Momjian
Vivek Khera wrote: BM == Bruce Momjian [EMAIL PROTECTED] writes: BM The 32k number is from Tatsuo testing a few years ago. Can you verify for me that these parameters in postgresql.conf are based on the BLCKSZ (ie one buffer is the size of the BLCKSIZ macro): shared_buffers

Re: [PERFORM] FreeBSD page size

2003-09-04 Thread Bruce Momjian
Vivek Khera wrote: BM == Bruce Momjian [EMAIL PROTECTED] writes: BM The 32k number is from Tatsuo testing a few years ago. Can you verify for me that these parameters in postgresql.conf are based on the BLCKSZ (ie one buffer is the size of the BLCKSIZ macro): shared_buffers

Re: [PERFORM] Query too slow

2003-09-04 Thread Bruce Momjian
-- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast

Re: [PERFORM] [GENERAL] how to get accurate values in pg_statistic

2003-09-07 Thread Bruce Momjian
lists at once with the unregister command (send unregister YourEmailAddressHere to [EMAIL PROTECTED]) -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ

Re: [PERFORM] Hardware recommendations to scale to silly load

2003-09-09 Thread Bruce Momjian
Inserts all in one transaction 3700 inserts/second Inserts all in one transaction 2500 inserts/second ECPG test program attached. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life

Re: [PERFORM] Hardware recommendations to scale to silly load

2003-09-10 Thread Bruce Momjian
on Inserts all in one transaction 3700 inserts/second Inserts in separate transactions870 inserts/second fsync off Inserts all in one transaction 3700 inserts/second Inserts in separate transactions 2500 inserts/second -- Bruce Momjian

Re: [PERFORM] [GENERAL] how to get accurate values in pg_statistic (continued)

2003-09-10 Thread Bruce Momjian
that are so bizarre that the planner still gets things wrong with a target of 1000, then I'd like to know more about why. Has there been any progress in determining if the number of default buckets (10) is the best value? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL

Re: [PERFORM] restore time: sort_mem vs. checkpoing_segments

2003-09-23 Thread Bruce Momjian
. Should that be decumented? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073

Re: [PERFORM] Tuning/performance issue...

2003-10-03 Thread Bruce Momjian
. * ---(end of broadcast)--- TIP 8: explain analyze is your friend -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13

Re: [PERFORM] Tuning/performance issue...

2003-10-03 Thread Bruce Momjian
Rod Taylor wrote: -- Start of PGP signed section. On Fri, 2003-10-03 at 21:39, Bruce Momjian wrote: I have updated the FAQ to be: In comparison to MySQL or leaner database systems, we are faster for multiple users, complex queries, and a read/write query load. MySQL

Re: [PERFORM] count(*) slow on large tables

2003-10-04 Thread Bruce Momjian
Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: Tom Lane wrote: ... and it would give the wrong answers. Unless the cache is somehow snapshot-aware, so that it can know which other transactions should be included in your count. The cache is an ordinary table, with xid's

Re: [PERFORM] COUNT(*) again (was Re: [HACKERS] Index/Function organized

2003-10-04 Thread Bruce Momjian
are sufficient reason why not, even before you look at the performance issues. What are the atomicity problems of adding a create/expire xid to the index tuples? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your

Re: [PERFORM] COUNT(*) again (was Re: [HACKERS] Index/Function organized

2003-10-05 Thread Bruce Momjian
Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: Tom Lane wrote: I think that's not happening, conditionally or otherwise. The atomicity problems alone are sufficient reason why not, even before you look at the performance issues. What are the atomicity problems of adding

Re: [PERFORM] reindex/vacuum locking/performance?

2003-10-06 Thread Bruce Momjian
. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end of broadcast

Re: [PERFORM] PostgreSQL vs. MySQL

2003-10-08 Thread Bruce Momjian
performance. /para -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073

Re: [PERFORM] PostgreSQL vs. MySQL

2003-10-08 Thread Bruce Momjian
? I think I can offer a 3-paragraph version which would cover the major points of setting PostgreSQL.conf. Yes, I think that is a good idea --- now, does it go in the install docs, or in the docs next to each GUC item? -- Bruce Momjian| http://candle.pha.pa.us

Re: [PERFORM] Sun performance - Major discovery!

2003-10-08 Thread Bruce Momjian
that did -O for any platforms that set no cflags, but I don't see that around anywhere. I recommend adding -O2, or at leaset -O to all these platforms --- we can then use platform testing to make sure they are working. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL

Re: [PERFORM] Sun performance - Major discovery!

2003-10-08 Thread Bruce Momjian
Tom Lane wrote: Neil Conway [EMAIL PROTECTED] writes: On Wed, 2003-10-08 at 14:31, Bruce Momjian wrote: Well, this is really embarassing. I can't imagine why we would not set at least -O on all platforms. I believe that autoconf will automatically select -O2 (when CFLAGS isn't already

Re: [HACKERS] [PERFORM] Sun performance - Major discovery!

2003-10-08 Thread Bruce Momjian
Peter Eisentraut wrote: Bruce Momjian writes: Well, this is really embarassing. I can't imagine why we would not set at least -O on all platforms. Looking at the template files, I see these have no optimization set: freebsd (non-alpha) I'm wondering what that had in mind

Re: [HACKERS] [PERFORM] Sun performance - Major discovery!

2003-10-08 Thread Bruce Momjian
Tom Lane wrote: Neil Conway [EMAIL PROTECTED] writes: On Wed, 2003-10-08 at 14:31, Bruce Momjian wrote: Well, this is really embarassing. I can't imagine why we would not set at least -O on all platforms. I believe that autoconf will automatically select -O2 (when CFLAGS isn't already

Re: [PERFORM] Sun performance - Major discovery!

2003-10-09 Thread Bruce Momjian
can get through to the mailing list cleanly -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania

Re: [PERFORM] Sun performance - Major discovery!

2003-10-09 Thread Bruce Momjian
What is the performance win for the -fast flag again? --- Jeff wrote: On Thu, 9 Oct 2003, Bruce Momjian wrote: So you want -fast added as default for non-gcc Solaris? You mentioned there is a warning generated

Re: [PERFORM] Sun performance - Major discovery!

2003-10-09 Thread Bruce Momjian
be better for this for now? I agree with the concern. I'd rather have slow'n'stable than fast-but-broken. FAQ added. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road

Re: [PERFORM] Sun performance - Major discovery!

2003-10-09 Thread Bruce Momjian
Jeff wrote: We're keeping the -O2 for gcc in the template and moving the mention of -fast to the FAQ, correct? gcc gets -O2, non-gcc gets -O, and -fast is in the FAQ, yea. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001

Re: [PERFORM] Any 7.4 w32 numbers in yet?

2003-10-09 Thread Bruce Momjian
We don't have it running yet. It will be running in 7.5. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square

Re: [PERFORM] Sun performance - Major discovery!

2003-10-09 Thread Bruce Momjian
Christopher Browne wrote: [EMAIL PROTECTED] (Bruce Momjian) writes: 5) How can I compile for optimum performance? Try using the -fast compile flag. The binaries might not be portable to other Solaris systems, and you might need to compile everything that links to PostgreSQL with -fast

Re: [PERFORM] further testing on IDE drives

2003-10-09 Thread Bruce Momjian
unregister YourEmailAddressHere to [EMAIL PROTECTED]) -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square

Re: [PERFORM] further testing on IDE drives

2003-10-10 Thread Bruce Momjian
scott.marlowe wrote: On Thu, 9 Oct 2003, Bruce Momjian wrote: scott.marlowe wrote: I was testing to get some idea of how to speed up the speed of pgbench with IDE drives and the write caching turned off in Linux (i.e. hdparm -W0 /dev/hdx). The only parameter that seems

Re: [PERFORM] further testing on IDE drives

2003-10-10 Thread Bruce Momjian
-based test machine here, and the TPCC databases. I would be interested to see if wal_sync_method = fsync is slower than wal_sync_method = open_sync. How often are we doing more then one write before a fsync anyway? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL

Re: [PERFORM] further testing on IDE drives

2003-10-10 Thread Bruce Momjian
Vivek Khera wrote: BM == Bruce Momjian [EMAIL PROTECTED] writes: Sounds reasonable to me. Are there many / any scenarios where a plain fsync would be faster than open_sync? BM Yes. If you were doing multiple WAL writes before transaction fsync, BM you would be fsyncing every write

Re: [PERFORM] go for a script! / ex: PostgreSQL vs. MySQL

2003-10-10 Thread Bruce Momjian
? Too many tunables are page dependant, which is infuriating when copying configs from DB to DB. I wish pgsql had some notion of percentages for values that end with a '%'. -sc Makes sense to me --- we already have some read-only GUC variables. -- Bruce Momjian| http

Re: [PERFORM] Sun performance - Major discovery!

2003-10-09 Thread Bruce Momjian
it be? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end

Re: [HACKERS] [PERFORM] Sun performance - Major discovery!

2003-10-13 Thread Bruce Momjian
Done as you suggested. --- Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: OK, patch attached and applied. It centralizes the optimization defaults into configure.in, rather than having CFLAGS= in the template

Re: [PERFORM] analyzing postgresql performance for dbt-2

2003-10-21 Thread Bruce Momjian
computer was saturated. Is there something I am missing? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square

Re: [PERFORM] RedHat Enterprise Linux ES 3 ?!?!

2003-10-23 Thread Bruce Momjian
for employing PostgreSQL folks --- here is a nice URL I read yesterday on the topic: http://news.com.com/2100-7344-5094774.html?tag=nl -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive

Re: [PERFORM] My own performance/tuning qa

2003-10-24 Thread Bruce Momjian
something wrong with the design. Yes, I think Tom's concern was priority inversion, where a low priority process holds a lock while a higher one waits for it. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life

Re: [PERFORM] analyzing postgresql performance for dbt-2

2003-10-25 Thread Bruce Momjian
that it appears something might be saturated, I just don't know where to look... Could the 10% be context switching time, or is the I/O saturated? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive

Re: [PERFORM] Adding foreign key performance

2003-10-28 Thread Bruce Momjian
(I think you have to drop the tables to erase the statistics). -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square

Re: [PERFORM] Adding foreign key performance

2003-10-28 Thread Bruce Momjian
depended on it ;o) Is there some way to name the foreign key so that it can be dropped later, or is there a way to drop the foreign key using information already in the database? You have to use ALTER TABLE DROP CONSTRAINT perhaps. -- Bruce Momjian| http

Re: [PERFORM] vacuum locking

2003-10-30 Thread Bruce Momjian
perspective. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073

Re: [PERFORM] Pg+Linux swap use

2003-11-06 Thread Bruce Momjian
isn't all that useful unless you are sure to keep all 4 cpu's busy. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown

Re: [PERFORM] IN surpasses NOT EXISTS in 7.4RC2 ??

2003-11-13 Thread Bruce Momjian
, NOT IN will often be faster that NOT EXISTS. NOT EXISTS didn't change --- there are restrictions on how far we can optimize NOT EXISTS. NOT IN has just become much faster in 7.4. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001

Re: [PERFORM] duration logging setting in 7.4

2003-11-17 Thread Bruce Momjian
: LOG: duration: 0.861 ms statement: select 1; -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania

Re: [PERFORM] duration logging setting in 7.4

2003-11-18 Thread Bruce Momjian
Wow, that is strange. If you don't use syslog, do you see the proper output? If you turn on log_statement, do you see the statements? --- Ryszard Lach wrote: On Mon, Nov 17, 2003 at 09:37:07PM -0500, Bruce Momjian wrote

Re: [PERFORM] duration logging setting in 7.4

2003-11-19 Thread Bruce Momjian
(not used by application explicite, probably added by JDBC driver), now they are missed - maybe this is the problem? No, those are missing because the new 7.4 wire protocol doesn't require those queries anymore --- the data is send automatically. -- Bruce Momjian| http

Re: [PERFORM] duration logging setting in 7.4

2003-11-19 Thread Bruce Momjian
Ryszard Lach wrote: On Wed, Nov 19, 2003 at 01:58:27PM -0500, Bruce Momjian wrote: Ryszard Lach wrote: There is another one thing: logs from the same database running on 7.3 and the same application contained lines like 'select getdatabaseencoding()', 'select datestyle

Re: [PERFORM] Is clustering possible to enhance the performance?

2003-12-01 Thread Bruce Momjian
. Dual cpu's allow multiple backends to use different cpu's, but a single session can't use more than one cpu. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road

Re: [PERFORM] A question on the query planner

2003-12-03 Thread Bruce Momjian
, there's something busted here. Will get back to you ... LOL, but I am not sure why. :-) -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup

Re: [PERFORM] autovacuum daemon stops doing work after about an

2003-12-07 Thread Bruce Momjian
)--- TIP 3: if posting/reading through Usenet, please send an appropriate subscribe-nomail command to [EMAIL PROTECTED] so that your message can get through to the mailing list cleanly -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED

[PERFORM] Update performance doc

2003-12-13 Thread Bruce Momjian
I have updated my hardware performance documentation to reflect the findings during the past few months on the performance list: http://candle.pha.pa.us/main/writings/pgsql/hw_performance/index.html Thanks. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL

Re: [PERFORM] optimizing Postgres queries

2004-01-05 Thread Bruce Momjian
to start to understand more about this optimization problem? Read the FAQ. There is an item about slow queries and indexes. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13

Re: [PERFORM] Postgres on Netapp

2004-01-23 Thread Bruce Momjian
a flag that allowed a database to be opened in read only mode without touching anything in the directory. PostgreSQL has to read the WAL to adjust the contents of the flat file on startup in such a setup, so I don't see how we could do it read-only. -- Bruce Momjian

Re: [PERFORM] High Performance/High Reliability File system on SuSE64

2004-01-31 Thread Bruce Momjian
that the postmaster did not still have some some data that was not flushed to disk yet. Uh, yea, it does. If the snapshot includes all of /data, including WAL/xlog, you can then back up the snapshot and restore it on another machine. It will restart just like a crash. -- Bruce Momjian

Re: [PERFORM] Disappointing performance in db migrated from MS SQL Server

2004-02-12 Thread Bruce Momjian
is that the tests are I/O bound and therefore the faster CPU's aren't helping PostgreSQL. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup

Re: [PERFORM] Feature request: smarter use of conditional indexes

2004-03-08 Thread Bruce Momjian
it appeared in the right place. That has gotten me a few times. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown

Re: [PERFORM] [ADMIN] syslog slowing the database?

2004-03-15 Thread Bruce Momjian
) with this command to start it in my rc.local file: su - postgres -c 'pg_ctl start | rotatelogs $PGDATA/pglog 86400 21' Sure, our documentation specifically mentions using rotatelogs. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-18 Thread Bruce Momjian
[ Attachment, skipping... ] -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-18 Thread Bruce Momjian
. Would other users run src/tools/fsync and report their findings so I can update the template files for their OS's? This is a process similar to our thread testing. Thanks. --- Bruce Momjian wrote: Mark Kirkwood wrote

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-18 Thread Bruce Momjian
Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: I have been poking around with our fsync default options to see if I can improve them. One issue is that we never default to O_SYNC, but default to O_DSYNC if it exists, which seems strange. As I recall, that was based on testing

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-18 Thread Bruce Momjian
Kurt Roeckx wrote: On Thu, Mar 18, 2004 at 01:50:32PM -0500, Bruce Momjian wrote: I'm not sure I believe these numbers at all... my experience is that getting trustworthy disk I/O numbers is *not* easy. These numbers were reproducable on all the platforms I tested. It's not because

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-18 Thread Bruce Momjian
Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: Tom Lane wrote: As I recall, that was based on testing on some different platforms. But why perfer O_DSYNC over fdatasync if you don't prefer O_SYNC over fsync? It's what tested out as the best bet. I think we were using

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-18 Thread Bruce Momjian
? http://www.postgresql.org/docs/faqs/FAQ.html -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-18 Thread Bruce Momjian
simple one. We don't need accurate numbers. We just need to know if fsync or O_SYNC is faster. 2) Won't Jan's work on 7.5 memory and I/O management mean that we have to re-evaluate synching anyway? No, it should not change sync issues. -- Bruce Momjian| http

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-18 Thread Bruce Momjian
that it works in a single process. Unix isn't very process-centered for I/O, so I don't think it would make much of a difference. Now, Win32, that might be an issue. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-22 Thread Bruce Momjian
and get some indication of how frequently writes and fsync's are going to WAL and modify tools/fsync to match what your DBT-2 test does. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-25 Thread Bruce Momjian
), and the access/modification timestamp is only in seconds, so I wonder of the OS only updates the inode once a second. What else would change in the inode more frequently than once a second? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED

Re: [PERFORM] [HACKERS] fsync method checking

2004-03-26 Thread Bruce Momjian
Odd. Which filesystem, which kernel? It seems fdatasync is broken and syncs the inode, too. It's linux-2.6.5-rc1 with ext2 filesystems. Would you benchmark open_sync for wal_sync_method too? -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED

Re: [PERFORM] good pc but bad performance,why?

2004-04-07 Thread Bruce Momjian
. Is this true? I was under the impression that without at least meta-data journaling postgresql could still be corrupted by power failure. It is false. ext2 isn't crash-safe, and PostgreSQL needs an intact file system for WAL recovery. -- Bruce Momjian| http

Re: [PERFORM] good pc but bad performance,why?

2004-04-07 Thread Bruce Momjian
Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: scott.marlowe wrote: There is no real need (or benefit) from having the database on a journalled filesystem - the journalling is only trying to give similar sorts of guarantees to what the fsync in PostgreSQL is doing

Re: [PERFORM] good pc but bad performance,why?

2004-04-07 Thread Bruce Momjian
Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: Tom Lane wrote: But it should be okay to set the filesystem to journal only its own metadata. There's no need for it to journal file contents. Can you set ext2 to journal metadata? I didn't know it could do that. No, ext2 has

Re: [PERFORM] good pc but bad performance,why?

2004-04-08 Thread Bruce Momjian
Andrew McMillan wrote: On Thu, 2004-04-08 at 14:13, Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: Tom Lane wrote: But it should be okay to set the filesystem to journal only its own metadata. There's no need for it to journal file contents. Can you set ext2 to journal

Re: [PERFORM] data=writeback

2004-04-08 Thread Bruce Momjian
continue. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup.| Newtown Square, Pennsylvania 19073 ---(end

Re: [PERFORM] Wierd context-switching issue on Xeon

2004-04-19 Thread Bruce Momjian
. I have BSD on a SuperMicro dual Xeon, so if folks want another hardware/OS combination to test, I can give out logins to my machine. http://candle.pha.pa.us/main/hardware.html -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610

Re: [PERFORM] Wierd context-switching issue on Xeon

2004-04-20 Thread Bruce Momjian
with peaks to 300k CS. Bummer.. Josh, I don't think you can bash the ServerWorks chipset here nor bigmem. Dave Cramer reproduced the problem on my SuperMicro dual Xeon on BSD/OS. -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610

[PERFORM] pgbench written in Pro*C

2004-04-20 Thread Bruce Momjian
I received a copy of pgbench rewritten in Pro*C, which is similar to embedded C. I think it was done so the same program could be tested on Oracle and PostgreSQL. Are folks interested in this code? Should it be put on gborg or in our /contrib/pgbench? -- Bruce Momjian

Re: [PERFORM] pgbench written in Pro*C

2004-04-20 Thread Bruce Momjian
Tom Lane wrote: Bruce Momjian [EMAIL PROTECTED] writes: I received a copy of pgbench rewritten in Pro*C, which is similar to embedded C. I think it was done so the same program could be tested on Oracle and PostgreSQL. Are folks interested in this code? Should it be put on gborg

Re: [PERFORM] Wierd context-switching issue on Xeon patch for 7.4.1

2004-04-21 Thread Bruce Momjian
) //if (sysctlbyname(hw.ncpu, ncpus, len, NULL, 0 ) == 0) cpucount = ncpus; } and I can confirm that on my computer it works: hw.ncpu = 2 -- Bruce Momjian| http://candle.pha.pa.us [EMAIL PROTECTED] | (610) 359-1001 + If your life

Re: [PERFORM] Wierd context-switching issue on Xeon patch for 7.4.1

2004-04-22 Thread Bruce Momjian
. Also, this is a fairly rare circumstance, I think; to judge by my clients, once a database server is in production nobody touches the hardware. A much simpler solution would be for the postmaster to run a test during startup. -- Bruce Momjian| http

  1   2   3   4   5   >