On Sun, 28 Nov 2004 17:53:38 -0500, Tom Lane wrote:
But is it cheaper, IO-wise to jump around in an index than to go back
and forth between index and tuple blocks?
Perhaps not --- but why would you be jumping around? Wouldn't the
needed info appear in consecutive locations in the index?
Troels Arvin [EMAIL PROTECTED] writes:
How much of[1] is still the case today?
Reference 1:
Stonebraker Olson: Large Object Support in POSTGRES (1993)
http://epoch.cs.berkeley.edu:8000/postgres/papers/S2K-93-30.pdf
Probably almost none of it ... the only thing I know about the
Berkeley-era
On Fri, 19 Nov 2004 10:35:20 -0500, Tom Lane wrote:
2. Does someone know of interesting documentation (perhaps
in the form of interesting code comments) which I should
read, as a basis for creating a non-standard index type
in PostgreSQL?
There's not a whole lot :-( and you should
Troels Arvin [EMAIL PROTECTED] writes:
What kind of (logical) block identifier should I point to in my index?
CTID (block # + line #) is the only valid pointer from an index to a
table. It doesn't change over the life of an index entry. I think
though that you'd be making a serious mistake by
On Sun, 28 Nov 2004 16:52:47 -0500, Tom Lane wrote:
CTID (block # + line #) is the only valid pointer from an index to a
table.
Thanks.
I think
though that you'd be making a serious mistake by not duplicating the
suffixes into the index (rather than expecting to retrieve them from the
Troels Arvin [EMAIL PROTECTED] writes:
On Sun, 28 Nov 2004 16:52:47 -0500, Tom Lane wrote:
You need to be able
to scan the index and identify rows matching a query without making lots
of probes into the table.
But is it cheaper, IO-wise to jump around in an index than to go back
and forth
Hello,
I'm working on a thesis project where I explore the addition of a
specialized, bioinformatics-related data type to a RDBMS. My choice of
RDBMS is PostgreSQL, of course, and I've started by adding a dnaseq (DNA
sequence) data type, using PostgreSQL's APIs for type additions.
The idea is to
Hi,
your project looks very attractive. In principle, suffix array should be
implemented using GiST framework. String Btree should be very useful
for your problem. My student is working on string btree library, but we
have no plan to intergrate it into postgresql.
Oleg
On Fri, 19 Nov 2004,
On R, 2004-11-19 at 12:42, Troels Arvin wrote:
The basic parts of the type are pretty much done. Those interested may
have a look at http://troels.arvin.dk/svn-snap/postgresql-dnaseq/ (the
code organization needs some clean-up). The basic type implementation
should be improved by adding more
Hi Troels,
This is not related to the database aspects of your question... But there
are more than 4 possible letters in DNA sequences, 16 in fact. Depending on
the accuracy of the DNA sequences you are storing, you may come across
ambiguity DNA bases, so your type will have to take these into
Hello Oleg,
On Fri, 2004-11-19 at 15:35 +0300, Oleg Bartunov wrote:
your project looks very attractive.
Thanks.
In principle, suffix array should be implemented using GiST framework.
But in a previous conversation between the two of us, you wrote that the
GiST wasn't suitable for this
On Fri, 19 Nov 2004 14:38:20 +0200, Hannu Krosing wrote:
Part of my current code concerns packing DNA characters: As the alphabet
of DNA strings is very small (four characters), it seems like a
straigt-forward optimization to store each character in two bits.
My advice would be to get it
On Fri, 19 Nov 2004, Troels Arvin wrote:
Hello Oleg,
On Fri, 2004-11-19 at 15:35 +0300, Oleg Bartunov wrote:
your project looks very attractive.
Thanks.
In principle, suffix array should be implemented using GiST framework.
But in a previous conversation between the two of us, you wrote that the
Troels Arvin [EMAIL PROTECTED] writes:
2. Does someone know of interesting documentation (perhaps
in the form of interesting code comments) which I should
read, as a basis for creating a non-standard index type
in PostgreSQL?
There's not a whole lot :-( and you should definitely
On Fri, 2004-11-19 at 10:42, Troels Arvin wrote:
Hello,
I'm working on a thesis project where I explore the addition of a
specialized, bioinformatics-related data type to a RDBMS. My choice of
RDBMS is PostgreSQL, of course, and I've started by adding a dnaseq (DNA
sequence) data type,
15 matches
Mail list logo
