Hi Jeff,
I've been assigned to do an initial review of your "new correlation
metric" patch.
If I'm grokking the thread, it looks like Tom suggested a substantial
change in the approach (targetting per-index correlation rather than
per-column) [1], and although you agreed with the spirit of his
su
On Mon, 2008-11-03 at 12:16 +1300, Mark Kirkwood wrote:
> Trying out a few different scenarios I ran across this:
>
> CONTEXT: xlog redo update: rel 1663/16384/16397; tid 9614/62; new 158828/59
> DEBUG: start recovery xid = 7002 lsn = 0/6F012EE4
> CONTEXT: xlog redo update: rel 1663/16384/163
On Sun, 2 Nov 2008, Josh Berkus wrote:
I'd start with command-line switches, e.g.
config --memory=32GB --type=DW --size=500GB --connections=20
That seems reasonable, I think I'll push a fancier UI on the backburner
then and just spec out an options interface like this one.
I think in initia
I met a problem with parsing Datum,
is there any good way to parse this
kind of value? Because I want to get
the value in an attribute of a table.
In TupleTableSlot, there is a tts_values
which is Datum type. Does anyone know how
to parse it?
Best Regards,
Zhe HE
TEL: (001) 646-789-3008
Address:96
Greg Smith <[EMAIL PROTECTED]> writes:
> On Sun, 2 Nov 2008, Tom Lane wrote:
>> The directions on the "move" tab imply that this is taken care of
>> automatically?
> What you get when you move something is redirect page that translates the
> old name into the new one. But if you were already tar
On Sun, 2 Nov 2008, Tom Lane wrote:
"Brendan Jurd" <[EMAIL PROTECTED]> writes:
Note that if you rename a page you also need to update any redirects
which point to it.
The directions on the "move" tab imply that this is taken care of
automatically?
What you get when you move something is red
I tried to apply this patch to CVS HEAD and it blew up all over the
place. It doesn't seem to be intended to apply against CVS HEAD; for
example, I don't have backend/access/heap/htup.c at all, so can't
apply changes to that file. I was able to clone the GIT repository
with the following command.
"Brendan Jurd" <[EMAIL PROTECTED]> writes:
> Note that if you rename a page you also need to update any redirects
> which point to it.
The directions on the "move" tab imply that this is taken care of
automatically? But anyway it'll just be a minor amount of extra
work when the time comes.
Josh Berkus <[EMAIL PROTECTED]> writes:
> Note that I'm going to propose replacing the wiki with real software
> come 8.5.
Where is this "real software" going to come from? I can't imagine that
software with more functionality than the wiki is going to spring into
existence without a huge invest
On Mon, Nov 3, 2008 at 2:31 PM, Tom Lane <[EMAIL PROTECTED]> wrote:
> The problem I'm having with this is that the current scheme for naming
> commitfest pages assumes you know the date for each one. It seems way
> premature to bet on when the first 8.5 fest will start.
>
> Wikimedia does seem to
Tom,
Thoughts, better ideas?
Nope, that's why I ignored CommitFestOpen. I couldn't figure out where
to point it.
Note that I'm going to propose replacing the wiki with real software
come 8.5. The wiki method confuses new reviewers, and makes me spend
about 500% as much time managing the
http://wiki.postgresql.org/wiki/CommitFest_2008-11
currently says that it's closed to new submissions, but this isn't
particularly helpful when the link it gives for where to submit them
points right back to itself. So, where should we make CommitFestOpen
redirect to?
The problem I'm having with
Here's an updated patch for just array_accum() with some simple docs. If
I should document this in more places, let me know.
I decided not to include array_agg() in this patch because it doesn't
support the standard's ORDER BY clause.
My reasoning is that, if someone is using the standard array_a
> From: Tom Lane [mailto:[EMAIL PROTECTED]
> What alternatives are there for people who do not run Windows?
>
> regards, tom lane
The TPC-H generator is a standard code base provided at
http://www.tpc.org/tpch/. We have been able to compile this code on
Linux.
However, we
"Lawrence, Ramon" <[EMAIL PROTECTED]> writes:
> The easiest way to test would be to generate your own TPC-H data and
> load it into a database for testing. I have posted the TPC-H generator
> at:
> http://people.ok.ubc.ca/rlawrenc/TPCHSkew.zip
> The generator can produce skewed data sets. It was
Martin Pihlak <[EMAIL PROTECTED]> writes:
> Attached is a patch that implements the described signalling. Additionally
> following non-related changes have been made:
> 1. fopen/fclose replaced with AllocateFile/FreeFile
> 2. pgstat_report_stat() now also checks if there are functions to report
> b
Greg Smith wrote:
While I've got a pretty clear vision for what I'm doing with this next
and will kick off a pgfoundry project real soon, I wanted to throw
this out as a WIP for feedback at this point. I was ultimately hoping
to one day have something like this shipped as a contrib/ module to
On Sun, Nov 2, 2008 at 4:48 PM, Lawrence, Ramon <[EMAIL PROTECTED]> wrote:
> Joshua,
>
> Thank you for offering to review the patch.
>
> The easiest way to test would be to generate your own TPC-H data and
> load it into a database for testing. I have posted the TPC-H generator
> at:
>
> http://pe
Joshua,
Thank you for offering to review the patch.
The easiest way to test would be to generate your own TPC-H data and
load it into a database for testing. I have posted the TPC-H generator
at:
http://people.ok.ubc.ca/rlawrenc/TPCHSkew.zip
The generator can produce skewed data sets. It was
I think the k hash functions are actually normally just different
slices of bits taken from one actual hash function anyways so it
sounds like you've done the right thing.
This sounds most interesting for multibatch hash joins if you could
build a bloom filter for the future batches to avoi
Trying out a few different scenarios I ran across this:
1/ Setup master and replica with replica using pg_standby
2/ Create a new database ("bench" in my case)
3/ Initialize pgbench schema size 100
4/ Run with 2 clients and 1 transactions
5/ Replica gets assertion failure
This is Postgres he
On Sun, Nov 2, 2008 at 5:36 PM, Hannes Eder <[EMAIL PROTECTED]> wrote:
> On Sun, Nov 2, 2008 at 10:49 PM, Jonah H. Harris <[EMAIL PROTECTED]> wrote:
>> Similarly, I
>> created a GUC to enable pruning, named bloom_pruning.
>
> I guess calls to bloom_filter_XXX should be surrounded by "if
> (bloom_pr
On Sun, Nov 2, 2008 at 10:49 PM, Jonah H. Harris <[EMAIL PROTECTED]> wrote:
> Similarly, I
> created a GUC to enable pruning, named bloom_pruning.
I guess calls to bloom_filter_XXX should be surrounded by "if
(bloom_pruning) ..." or a similar construct, i.e. make use of the GUC
variable bloom_prun
All,
Attached is an initial patch I've been playing with which uses Bloom
filters to reduce unnecessary processing of outer tuples in hash
joins. In short, this works by creating a Bloom filter, adding all
relevant tuples for the inner relation, and querying the filter (for
existence) when retrie
Kris Jurka <[EMAIL PROTECTED]> writes:
> Since we require every process to map the shared memory region to the same
> address, we don't need the MAKE_PTR/OFFSET code that was needed when that
> was not the case. This patch makes shared memory pointers just like
> regular pointers.
Applied with
Simon Riggs wrote:
On Sat, 2008-11-01 at 15:18 +0200, Heikki Linnakangas wrote:
Simon Riggs wrote:
On Fri, 2008-10-31 at 15:05 +, Heikki Linnakangas wrote:
Log Message:
---
Unite ReadBufferWithFork, ReadBufferWithStrategy, and ZeroOrReadBuffer
functions into one ReadBufferExtended
Tom Lane napsal(a):
Heikki Linnakangas <[EMAIL PROTECTED]> writes:
Hmm, you're right. I think it can be made to work by storing the *end*
offset of each chunk. To find the chunk containing offset X, search for
the first chunk with end_offset > X.
Yeah, that seems like it would work, and it wo
Hitoshi Harada Wrote:
> Thanks for your test. Didn't post publicly, I've also tested real
> problems and performed better than I thought. If you can afford it,
> could you add selfjoin cases? It's like:
Ok, did self joins with some. I don't know if it's possible with all.
Test Sub query Self jo
* Tom Lane ([EMAIL PROTECTED]) wrote:
> Stephen Frost <[EMAIL PROTECTED]> writes:
> > ... A case I just realized might be an issue is
> > doing a 'select 1 from x;' where you have *no* rights on x, or any
> > columns in it, would still get you the rowcount.
>
> Well, if you have table-level select
* Tom Lane ([EMAIL PROTECTED]) wrote:
> Stephen Frost <[EMAIL PROTECTED]> writes:
> > ... A case I just realized might be an issue is
> > doing a 'select 1 from x;' where you have *no* rights on x, or any
> > columns in it, would still get you the rowcount.
>
> Well, if you have table-level select
On Sat, Nov 01, 2008 at 08:07:13PM -0400, Andrew Sullivan wrote:
> On Sat, Nov 01, 2008 at 04:21:30PM -0400, Robert Haas wrote:
>
> > It looks to me like there are at least half a dozen patches submitted
> > in the last week that are pretty half-baked and fall into the category
> > of "Let's submi
Sent from my iPhone
On 2 Nov 2008, at 06:57, "Joshua D. Drake" <[EMAIL PROTECTED]> wrote:
Josh Berkus wrote:
Tom, Robert, Simon,
What, are people just on edge because of the US election?
It looks to me like the commitfest system is going really well. Of
course, we'll see how long it takes
Josh Berkus wrote:
Tom, Robert, Simon,
What, are people just on edge because of the US election?
It looks to me like the commitfest system is going really well. Of
course, we'll see how long it takes to close out 8.4. But I think we're
in much better shape than we were for 8.3. We're even
33 matches
Mail list logo
