ed to be a completely database agnostic
> tpc-c like java based benchmark.
With the exception that it analyzes Postgres tables but not Oracle or
InnoDB, I agree with that. The goal of BenchmarkSQL was to be a database
agnostic benchmark kit.
--
Jonah H. Harris, Senior DBA
myYearbook.com
rk. If Sergio wants to correct them and/or
qualify them, that's cool with me. I just don't like people relying on
questionable and/or unclear data.
--
Jonah H. Harris, Senior DBA
myYearbook.com
On Fri, Feb 20, 2009 at 2:48 PM, Jonah H. Harris wrote:
> Having this said, the benchmark is not as unfair as you thought. I've
>> taken care to prepare all databases to meet similar values for their
>> cache, buffers and I/O configuration (to what's possible given their
sized
logs/log buffers makes a big difference. There are also several other
concurrency-related tunables which contribute to it as well.
--
Jonah H. Harris, Senior DBA
myYearbook.com
of the box in OLTP
> config) will come out 60%" or "Oracle comes out twice as fast as PG on
> Linux" without any proof to support this words. At least, benchmarks
> are refutable by using logic.
Your benchmark was flawed, you didn't tune correctly, and you made
statements based on bad data; refute that logic :)
--
Jonah H. Harris, Senior DBA
myYearbook.com
ming benchmarks, there are a lot of things to take into
consideration. If you're just performing out-of-the-box tests, then that's
fine, but you have to make sure the benchmark kit doesn't optimize itself
for any one of those databases (which it does for PG).
--
Jonah H. Harris, Senior DBA
myYearbook.com
eries. After you resolve that, you'll quickly notice
that Postgres' buffer manager design and the lack of a good
multi-block read quickly comes into play. The hash join
implementation also has a couple issues which I've recently seen
mentioned in other threads.
Use DBT-3, it will
munity isn't going
to see it as valuable enough to add to the core engine. IIRC,
systemtap is pretty much dead :(
--
Jonah H. Harris, Senior DBA
myYearbook.com
--
Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-performance
On Mon, Jul 7, 2008 at 9:23 AM, Merlin Moncure <[EMAIL PROTECTED]> wrote:
> I have a lot of problems with your statements. First of all, we are
> not really talking about 'RAM' storage...I think your comments would
> be more on point if we were talking about mounting database storage
> directly fr
; page ??
IIRC, I don't think so. I think you'd have to u se something like
pg_filedump to see if you have rows migrated to other blocks due to
updates.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
499 Thornall Street, 2nd
pdate on the same page.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECTED]
Edison, NJ 08837 | http://www.enterprisedb.com/
--
Sent via pgsql-performance mailing list (pgsql-performance@postgresq
On Wed, Jul 2, 2008 at 8:31 AM, Gauri Kanekar
<[EMAIL PROTECTED]> wrote:
> Performance of Hot was much better on 30June as compared to 2nd July.
Did you happen to VACUUM FULL or CLUSTER anything?
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporat
an utilize the CPU's free
cycles (during the wait) to handle other users. In short, sometimes,
disk I/O is a good thing; it just depends on what you need.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
499 Thornall Street, 2nd
As usual, it really just depends on the application and its requirements.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECTED]
Edison, NJ 08837 | http://www.enterprisedb.com/
--
Sent via pgsql-per
QL a federated database. The pl/proxy
architecture certainly doesn't resemble federated in the sense of the
other database vendors.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECTED]
Edis
Also check out Skytools: http://skytools.projects.postgresql.org/doc/
Hmm, I didn't think the Skype tools could really provide federated
database functionality without a good amount of custom work. Or, am I
mistaken?
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fa
B includes
Oracle-style database links (SELECT col FROM [EMAIL PROTECTED]) which support
predicate push-down.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECTED]
Edison, NJ 08837 | http
ta.
Can't argue with that one.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECTED]
Edison, NJ 08837 | http://www.enterprisedb.com/
--
Sent via pgsql-performance mailing list (pgsql-
cument containing the complexity of each
aggregate, but it's sometimes left as a comment in the souce code.
IIRC, COUNT (non-distinct) is currently O(n), where n also includes
evaluation of tuples not represented in the final count (due to
Postgres' MVCC design).
--
Jonah H. Harris,
On Sat, May 10, 2008 at 5:12 PM, Tom Lane <[EMAIL PROTECTED]> wrote:
> "Jonah H. Harris" <[EMAIL PROTECTED]> writes:
>> Repost to -hackers, you're more likely to get a response on this topic.
>
> Probably not, unless you cite a more readily available
But seems like the thread is
> loosed in tonn of other threads.
>
> --
> Sent via pgsql-performance mailing list (pgsql-performance@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-performance
>
--
Jonah H. Harris, Sr. Software
On Mon, Apr 28, 2008 at 5:37 PM, Adonias Malosso <[EMAIL PROTECTED]> wrote:
> Thank you for the answer. Good to know about this enterprise DB feature.
No problem.
> I´ll follow using pgloader.
That's fine. Though, I'd really suggest pg_bulkload, it's quite a bit f
h
also saves you the intermediate step of dumping the data.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation | fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECTED]
Edison, NJ 08837 | http://www.enterprisedb.com/
--
Sent via pgsql-performance
On Jan 10, 2008 6:25 PM, Steve Atkins <[EMAIL PROTECTED]> wrote:
> http://pgfoundry.org/projects/ip4r/
>
> That has the advantage over using integers, or the built-in inet type,
> of being indexable for range and overlap queries.
Agreed. ip4r is da bomb.
--
Jonah H. Ha
secure, and as JD always says, it just makes it difficult for the
average user to understand what it's doing.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECT
On Dec 14, 2007 2:03 PM, Bill Moran <[EMAIL PROTECTED]> wrote:
> I disagree here. If they're connecting remotely to PG, they have no
> direct access to the disk.
pg_read_file?
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation
parallel index builds or
other useful maintenance features... but it can do fairly good query
result parallelization.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499 Thornall Street, 2nd Floor
scans were the only thing
it could be useful for.
> Could someone shed some light on the current or future abilities of PG for
> making use of multiple cores to execute a single query?
Currently, the only way to parallelize a query in Postgres is to use pgpool-II.
http://pgpool.projects.postgre
On Nov 22, 2007 10:45 AM, Kevin Grittner <[EMAIL PROTECTED]> wrote:
> I suggest testing with some form of connection pooling.
Yeah, that's one of the reasons I suggested DBT-2. It pools
connections and is the most mature TPC-C-like test for Postgres.
--
Jonah H. Harris, Sr. Soft
commits are
not guaranteed (unlike your Oracle configuration). If you want
apples-to-apples, you need to turn fsync on.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTE
the outer table is
larger than the inner table, or the inner table itself is overly
large).
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECTED]
Edison, NJ 088
On Nov 16, 2007 10:56 AM, Dave Dutcher <[EMAIL PROTECTED]> wrote:
> I don't know about that. There are times when it is the right plan:
Agreed. IMHO, there's nothing wrong with nested-loop join as long as
it's being used properly.
--
Jonah H. Harris, Sr. So
D advocacy, mostly :) ), it contains a direct
> comparison between MySQL and PostgreSQL on various platforms, with
> PostgreSQL winning!
:)
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499 Thornall Street, 2nd
ze at runtime instead of having to batch this huge update?
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499 Thornall Street, 2nd Floor | [EMAIL PROTECTED]
Edison, NJ 08837| http://www.ent
cles where articles.article_id =
> links.article_to)
try:
UPDATE links
SET target_size = size
FROM articles
WHERE articles.article_id = links.article_to;
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499 Thornall St
= $2
AND calllog_self < $3
OR calllog_mainteng = $1
AND calllog_phase < 8
ORDER BY calllog_mainteng DESC,
calllog_phase DESC,
calllog_self DESC limit 25;
postgres# EXECUTE yourplan('124 ', 8, 366942);
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
it is still performing additional I/Os and additional CPU work to
read bloated data.
> Am I understanding it right?
Now, I think so.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499 Thornall Street,
sequentially, are performing
twice as many I/Os to do so. Which means you're actually waiting on
two things, I/O and additional CPU time reading blocks that have very
little viable data in them.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation
future, you probably want to set fillfactor
to a reasonable amount to account for updates-to-blocks-between-vacuum
to try and capture as few row-migrations as possible.
--
Jonah H. Harris, Sr. Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
499
ly on the OS to handle process
management across multiple CPUs/cores.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin, New Jersey 08830| http://www.ente
On 7/18/07, Benjamin Arai <[EMAIL PROTECTED]> wrote:
But I want to parrallelize searches if possible to reduce
the perofrmance loss of having multiple tables.
PostgreSQL does not support parallel query. Parallel query on top of
PostgreSQL is provided by ExtenDB and PGPool-II.
--
J
most of my tests. But, I haven't
done significant testing with direct I/O lately.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin, New Jersey 08
On 7/9/07, Joshua D. Drake <[EMAIL PROTECTED]> wrote:
PostgreSQL still beats MySQL ;)
Agreed.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin, New Jersey
es. A few other small differences as well if you dig into the
configurations, all of which I noted favored the PG system.
Agreed.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL P
On 7/9/07, Jignesh K. Shah <[EMAIL PROTECTED]> wrote:
I think this result will be useful for performance discussions of
postgresql against other databases.
I'm happy to see an industry-standard benchmark result for PostgreSQL.
Great work guys!
--
Jonah H. Harris, Software Archit
e list?
Yeah, Josh B. asked it to be toned down to the original list which
should've been involved. Which I think should be pgsql-admin or
pgsql-advocacy... your thoughts?
I think the Oracle discussion is over, David T. just needs URL references IMHO.
--
Jonah H. Harris, Software Archit
ng the topic again? My vote is for
the latter; it served no purpose other than to push the
competitiveness topic again.
I haven't seen any bashing going on yet. Shall we start with the closed
mindedness and unfairness of per cpu license and support models?
Not preferably, you make me type
-list gospel.
All of us have noticed the anti-MySQL bashing based on problems with
MySQL 3.23... Berkus and others (including yourself, if I am correct),
have corrected people on not making invalid comparisons against
ancient versions. I'm only doing the same where Oracle, IBM, and
Mic
On 6/18/07, Andreas Kostyrka <[EMAIL PROTECTED]> wrote:
As a cynic, I might ask, what Oracle is fearing?
As a realist, I might ask, how many times do we have to answer this
type of anti-commercial-database flamewar-starting question?
--
Jonah H. Harris, Software Architect |
comparison.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin, New Jersey 08830| http://www.enterprisedb.com/
---(end o
se
significantly different algorithms and optimizations. Likewise, there
is more tuning that can be done with Oracle given the amount of time
and money one has to spend on it. Again, cost/benefit analysis on
this type of an issue... but you're right, there is no "magic cure".
--
J
rite horror story about the grotty corners of
that software;
Of course, but they also never say why it was caused. With Oracle,
almost all bad-performance cases I've seen are related to improper
tuning and/or hardware; even by experienced DBAs.
--
Jonah H. Harris, Software Architect | pho
I usually encourage such people actually to perform the
analysis of the license, salary, contingency, and migrations costs
Yes, this is the best way.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor
gainst.
This is nothing against you, but it always starts an avalanche of,
"look how perfect we are compared to everyone else."
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor|
erts tuning the database incorrectly, writing a benchmark paper
about it, and making the software look bad.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin
MD1 ...
D5 + D6 = MD2 ...
MD0 + MD1 + MD2 = MDF (RAID 0)
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin, New Jersey 08830| http:/
when your own analysis is somewhat flawed.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin, New Jersey 08830| http://www.enterprisedb.com/
--
or speeding this up a
bit wouldn't we?
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin, New Jersey 08830| http://www.enterprisedb.com/
--
ically. You should look into pgpool or connection pooling from
the application.
--
Jonah H. Harris, Software Architect | phone: 732.331.1324
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 3rd Floor| [EMAIL PROTECTED]
Iselin, New Jersey 08830
and tracer available.
--
Jonah H. Harris, Software Architect | phone: 732.331.1300
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 2nd Floor| [EMAIL PROTECTED]
Iselin, New Jersey 08830| http://www.enterprisedb.com/
---(end of
7;t think you can point to any specific design
feature and say it's essential just on the basis of
bottom-line results. You have to look at the actual
benefit the specific wins.
True.
--
Jonah H. Harris, Software Architect | phone: 732.331.1300
EnterpriseDB Corporation
nice
happy medium for us.
/me waits for the obligatory and predictable, "the benchmarks are
flawed" response.
--
Jonah H. Harris, Software Architect | phone: 732.331.1300
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 2nd Floor| [EMAIL PROTEC
On 6/15/06, Jonah H. Harris <[EMAIL PROTECTED]> wrote:
On 6/15/06, Dan Gorman <[EMAIL PROTECTED]> wrote:
> shelfs. Any generic advice other than the NetApp (their NFS oracle
> tuning options) that might be useful? (e.g. turning off snapshots)
I was using PostgreSQL on a 980c
application are you running? OLTP? If so, what type of
transaction volume? Are you planning to use any Flex* or Snap*
features? What type of volume layouts are you using?
--
Jonah H. Harris, Software Architect | phone: 732.331.1300
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood
pp, this is
correct. fsync should be turned on, but you will not incur the *real*
direct-to-disk cost of the sync, it will be direct-to-NVRAM.
--
Jonah H. Harris, Software Architect | phone: 732.331.1300
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 2nd Floor| [EMAI
if It
exist then the postgres must do an update instead an insert.
PostgreSQL does not support MERGE at the moment, sorry.
--
Jonah H. Harris, Software Architect | phone: 732.331.1300
EnterpriseDB Corporation| fax: 732.331.1301
33 Wood Ave S, 2nd Floor| [EMAIL PROTECTED]
p://archives.postgresql.org-- Jonah H. Harris, Database Internals Architect
EnterpriseDB Corporation732.331.1324
67 matches
Mail list logo
