GSoC 2014 - mentors, students and admins

On Tue, Jan 28, 2014 at 11:04 PM, Thom Brown <thom@linux.com> wrote:

Hi all,

Application to Google Summer of Code 2014 can be made as of next
Monday (3rd Feb), and then there will be a 12 day window in which to
submit an application.

I'd like to gauge interest from both mentors and students as to
whether we'll want to do this.

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

Hi,

I would like to bring up the addition to MADLIB algorithms again this year.

Also, some work on the foreign table constraints could be helpful.

Regards,

Atri

--
Regards,

Atri
*l'apprenant*

On Tue, Jan 28, 2014 at 11:16 PM, Atri Sharma <atri.jiit@gmail.com> wrote:

On Tue, Jan 28, 2014 at 11:04 PM, Thom Brown <thom@linux.com> wrote:

Hi all,

Application to Google Summer of Code 2014 can be made as of next
Monday (3rd Feb), and then there will be a 12 day window in which to
submit an application.

I'd like to gauge interest from both mentors and students as to
whether we'll want to do this.

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

Hi,

I would like to bring up the addition to MADLIB algorithms again this year.

Also, some work on the foreign table constraints could be helpful.

Hi,

Also, can we consider a project in an extension to be a project in GSoC
2014 as GSoC 2014 under PostgreSQL?

I was thinking of having some support for writable FDW in JDBC_FDW, if
possible.

Regards,

Atri
--
Regards,

Atri
*l'apprenant*

On 01/28/2014 09:46 AM, Atri Sharma wrote:

I would like to bring up the addition to MADLIB algorithms again this year.

Also, some work on the foreign table constraints could be helpful.

We can only take MADLIB this year if we have confirmed mentors who are
MADLIB committers before the end of the application period (Feb 15). We
can't have a repeat of last year.

--
Josh Berkus
PostgreSQL Experts Inc.
http://pgexperts.com

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

Thom,

* Thom Brown (thom@linux.com) wrote:

Application to Google Summer of Code 2014 can be made as of next
Monday (3rd Feb), and then there will be a 12 day window in which to
submit an application.

This is just for PG to be a participating organization, right? There's
a while before mentors and students get invovled, as I understand it.

I'd like to gauge interest from both mentors and students as to
whether we'll want to do this.

Yes.

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Having you do it works for me. :)

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

I'm interested in mentoring and, unlike previous years, I've been
collecting a personal list of things that I'd like to see worked on for
PG which could be GSoC projects and will provide such in the next few
days to this list (unless there's a different list that people want such
posted to..?).

Thanks,

Stephen

Hi all,

I'd very interesting in taking part in the GSoC with PostgreSQL, as a
student.
Being quite busy at the moment (exam time), I haven't had time to think of
a subject yet, even though I could see some topics I found interesting.

2014-01-28 Josh Berkus <josh@agliodbs.com>

On 01/28/2014 09:46 AM, Atri Sharma wrote:

I would like to bring up the addition to MADLIB algorithms again this

year.
We can only take MADLIB this year if we have confirmed mentors who are
MADLIB committers before the end of the application period (Feb 15). We
can't have a repeat of last year.

Indeed, I've been unlucky last year when I tried to apply to a subject with
Madlib, mostly due to their lack of reactivity. Even though their subjects
interest me, I don't intend to try my luck again.

Atri Sharma said:

Also, some work on the foreign table constraints could be helpful.

What kind of work do you have in mind?

Stephen Frost said:

I'm interested in mentoring and, unlike previous years, I've been
collecting a personal list of things that I'd like to see worked on for
PG which could be GSoC projects and will provide such in the next few
days to this list (unless there's a different list that people want such
posted to..?).

IMO, the best place would be the wiki page for GSoC (
http://wiki.postgresql.org/wiki/GSoC_2014), it would avoid interested
students (including me :) ) having to look for possible subjects in lots of
different places.

Regards,
Maxence
--
Maxence Ahlouche
06 06 66 97 00

On 28 January 2014 19:43, Stephen Frost <sfrost@snowman.net> wrote:

Thom,

* Thom Brown (thom@linux.com) wrote:

Application to Google Summer of Code 2014 can be made as of next
Monday (3rd Feb), and then there will be a 12 day window in which to
submit an application.

This is just for PG to be a participating organization, right? There's
a while before mentors and students get invovled, as I understand it.

Yes, correct. Students and mentors don't need to be signed up until April.

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

I'm interested in mentoring and, unlike previous years, I've been
collecting a personal list of things that I'd like to see worked on for
PG which could be GSoC projects and will provide such in the next few
days to this list (unless there's a different list that people want such
posted to..?).

That's great. I don't see any problem with posting suggestions here,
although I'd suggest refraining from going in-depth as that can come
later. If there's enough interest and agreement, we'll go ahead and
apply.

Thom

--
Sent via pgsql-advocacy mailing list (pgsql-advocacy@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-advocacy

On 01/28/2014 07:34 PM, Thom Brown wrote:

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Please do, thanks!

Who would be up for mentoring this year?

I can mentor.

- Heikki

--
Sent via pgsql-advocacy mailing list (pgsql-advocacy@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-advocacy

On Tue, Jan 28, 2014 at 05:34:21PM +0000, Thom Brown wrote:

Hi all,

Application to Google Summer of Code 2014 can be made as of next
Monday (3rd Feb), and then there will be a 12 day window in which to
submit an application.

I'd like to gauge interest from both mentors and students as to
whether we'll want to do this.

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Thanks for your hard work administering last year, and thanks even
more for taking this on in light of that experience :)

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

I'd be delighted to mentor.

Cheers,
David.
--
David Fetter <david@fetter.org> http://fetter.org/
Phone: +1 415 235 3778 AIM: dfetter666 Yahoo!: dfetter
Skype: davidfetter XMPP: david.fetter@gmail.com
iCal: webcal://www.tripit.com/feed/ical/people/david74/tripit.ics

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate

--
Sent via pgsql-advocacy mailing list (pgsql-advocacy@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-advocacy

Hi!

On Tue, Jan 28, 2014 at 9:34 PM, Thom Brown <thom@linux.com> wrote:

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Thanks for your work. I would like to see you as admin this year again.

Who would be up for mentoring this year? And are there any project

ideas folk would like to suggest?

I would like to be mentor.

------
With best regards,
Alexander Korotkov.

Hi,

On 01/28/2014 06:46 PM, Atri Sharma wrote:

On Tue, Jan 28, 2014 at 11:04 PM, Thom Brown <thom@linux.com
<mailto:thom@linux.com>> wrote:

Hi all,

Application to Google Summer of Code 2014 can be made as of next
Monday (3rd Feb), and then there will be a 12 day window in which to
submit an application.

I'd like to gauge interest from both mentors and students as to
whether we'll want to do this.

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

I would like to bring up the addition to MADLIB algorithms again this year.

I've spoken with the MADlib team at goivotal and they are ok to support
this proposal. Therefore I offer to mentor this.

Regards,

--
Andreas 'ads' Scherbaum
German PostgreSQL User Group
European PostgreSQL User Group - Board of Directors
Volunteer Regional Contact, Germany - PostgreSQL Project

--
Sent via pgsql-advocacy mailing list (pgsql-advocacy@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-advocacy

Awesome. I can be an assistant mentor for this one is possible or I could
mentor some other project.

On Tuesday, February 25, 2014, Andreas 'ads' Scherbaum <
adsmail@wars-nicht.de> wrote:

Hi,

On 01/28/2014 06:46 PM, Atri Sharma wrote:

On Tue, Jan 28, 2014 at 11:04 PM, Thom Brown <thom@linux.com
<mailto:thom@linux.com>> wrote:

Hi all,

Application to Google Summer of Code 2014 can be made as of next
Monday (3rd Feb), and then there will be a 12 day window in which to
submit an application.

I'd like to gauge interest from both mentors and students as to
whether we'll want to do this.

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

I would like to bring up the addition to MADLIB algorithms again this
year.

I've spoken with the MADlib team at goivotal and they are ok to support
this proposal. Therefore I offer to mentor this.

Regards,

--
Andreas 'ads' Scherbaum
German PostgreSQL User Group
European PostgreSQL User Group - Board of Directors
Volunteer Regional Contact, Germany - PostgreSQL Project

--
Regards,

Atri
*l'apprenant*

On 25 February 2014 13:28, Andreas 'ads' Scherbaum <adsmail@wars-nicht.de>wrote:

Hi,

On 01/28/2014 06:46 PM, Atri Sharma wrote:

On Tue, Jan 28, 2014 at 11:04 PM, Thom Brown <thom@linux.com
<mailto:thom@linux.com>> wrote:

Hi all,

Application to Google Summer of Code 2014 can be made as of next
Monday (3rd Feb), and then there will be a 12 day window in which to
submit an application.

I'd like to gauge interest from both mentors and students as to
whether we'll want to do this.

And I'd be fine with being admin again this year, unless there's
anyone else who would like to take up the mantle?

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

I would like to bring up the addition to MADLIB algorithms again this
year.

I've spoken with the MADlib team at goivotal and they are ok to support
this proposal. Therefore I offer to mentor this.

Are there any more prospective mentors? We'll want some folk to act as
back-up mentors too to ensure projects can still be completed should any
mentor become unavailable.

--
Thom

On Thu, Feb 27, 2014 at 07:54:13PM +0000, Thom Brown wrote:

On 25 February 2014 13:28, Andreas 'ads' Scherbaum <adsmail@wars-nicht.de>wrote:

I've spoken with the MADlib team at goivotal and they are ok to
support this proposal. Therefore I offer to mentor this.

Are there any more prospective mentors? We'll want some folk to act
as back-up mentors too to ensure projects can still be completed
should any mentor become unavailable.

For MADlib, no. Are you asking for mentors in general?

Cheers,
David.
--
David Fetter <david@fetter.org> http://fetter.org/
Phone: +1 415 235 3778 AIM: dfetter666 Yahoo!: dfetter
Skype: davidfetter XMPP: david.fetter@gmail.com
iCal: webcal://www.tripit.com/feed/ical/people/david74/tripit.ics

Remember to vote!
Consider donating to Postgres: http://www.postgresql.org/about/donate

--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers

On 27 February 2014 21:08, David Fetter <david@fetter.org> wrote:

On Thu, Feb 27, 2014 at 07:54:13PM +0000, Thom Brown wrote:

On 25 February 2014 13:28, Andreas 'ads' Scherbaum <

adsmail@wars-nicht.de>wrote:

I've spoken with the MADlib team at goivotal and they are ok to
support this proposal. Therefore I offer to mentor this.

Are there any more prospective mentors? We'll want some folk to act
as back-up mentors too to ensure projects can still be completed
should any mentor become unavailable.

For MADlib, no. Are you asking for mentors in general?

Ah yes, I should clarify. Yes, mentors in general.

--
Thom

On Tue, Jan 28, 2014 at 5:34 PM, Thom Brown <thom@linux.com> wrote:

Who would be up for mentoring this year? And are there any project
ideas folk would like to suggest?

I mentored in the past and felt I didn't do a very good job because I
didn't really understand the project the student was working on.

There's precisely one project that I feel I would be competent to
mentor at this point. Making hash indexes WAL recoverable. This is
something that's easy to define the scope of and easy to determine if
the student is on track and easy to measure when finished. It's
something where as far as I can tell all the mentor work will be
purely technical advice.

Also it's something the project really really needs and is perfectly
sized for a GSOC project IMHO. Also it's a great project for a student
who might be interested in working on Postgres in the future since it
requires learning all our idiosyncratic build and source conventions
but doesn't require huge or controversial architectural changes.

I fear a number of items in the Wiki seem unrealistically large
projects for GSOC IMNSHO.

--
greg

--
Sent via pgsql-advocacy mailing list (pgsql-advocacy@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-advocacy

W dniu 27.02.2014 22:25, Thom Brown pisze:

On 27 February 2014 21:08, David Fetter <david@fetter.org
<mailto:david@fetter.org>> wrote:

For MADlib, no. Are you asking for mentors in general?

Ah yes, I should clarify. Yes, mentors in general.

In general I can help but I'm not sure if I'm not too fresh in pgsql ;)
However after GSOC as student I can try "the another side".
Regards,
Karol

Hi Greg, pgsql-advocacy, and pgsql-hackers,

I'm interested in doing my GSoC project on this idea. I'm new to indexing and WAL, which I haven't encountered in my classes, but it sounds interesting and valuable to Postgresql. So here's my draft proposal. Do you mind giving your opinion and corrections? With your help I'll add some technical detail to my plans.

Thanks,
Tan Tran

Introduction
In write-ahead logging (WAL), all modifications to a database are written to a write-ahead log before being flushed to disk at periodic checkpoints. This method saves I/O operations, enables a continuous backup, and, in the case of database failure, guarantees data integrity up until the last saved checkpoint. In Postgresql’s implementation, transactions are written to XLog, which is divided into 16MB files (“segments”) that together comprise a complete history of transactions. Transactions are continually appended to the latest segment, while checkpointing continually archives segments up until the last checkpoint. Internally, a suite of XLog structures and functions interfaces with the various resource managers so they can log a sufficient amount of data to restore data (“redo”) in case of failure.
Another Postgresql feature is the creation of indexes on a invariant custom field; for example, on the LastName of a Person even though the primary key is ID. These custom indexes speed up row lookup. Postgres currently supports four index types: B-tree, GiST, and GIN, and hash. Indexes on the former three are WAL-recoverable, but hashing is not.

2. Proposal
As a GSoC student, I will implement WAL recovery of hash indexes using the other index types’ WAL code as a guide. Roughly, I will:
- Devise a way to store and retrieve hashing data within the XLog data structures.
- In the existing skeleton for hash_redo(XLogRecPtr lsn, XLogRecord *record) in hash.c, branch to code for the various redo operations: creating an index, inserting into an index, deleting an index, and page operations (split, delete, update?).
- Code each branch by drawing on examples from btree_redo, gin_redo, and gist_redo, the existing XLog code of the other index types.

Benefits
Hash index searching is O(1), which is asymptotically faster than the O(n lg n) searching of a B-tree, and does not require custom indexing functions like GIN and GIST inherently do. Therefore it is desirable for rows that will only be retrieved on an equality or inequality relation. However, two things currently stand in the way of its popular use. From the Postgresql documentation,
“Hash index operations are not presently WAL-logged, so hash indexes might need to be rebuilt with REINDEX after a database crash if there were unwritten changes. Also, changes to hash indexes are not replicated over streaming or file-based replication after the initial base backup, so they give wrong answers to queries that subsequently use them. For these reasons, hash index use is presently discouraged.”
My project would solve the first problem, after which I would like to stay on and fix the second.

To be written: Quantifiable Results, Schedule, Completeness Criteria, Bio

On Feb 28, 2014, at 6:21 AM, Greg Stark <stark@mit.edu> wrote:

Earlier I posted an email to this thread that I realize "hijacked" the discussion. Please continue replying to here instead.

On Feb 28, 2014, at 6:59 AM, Karol Trzcionka <karlikt@gmail.com> wrote:

Hi all,

Earlier I posted this in the wrong thread. Please excuse the double posting.

Tan Tran

Begin forwarded message: