Re: Guidelines for GSoC student proposals / EliminateO(N^2) scaling from rw-conflict tracking in serializable transactions - Mailing list pgsql-hackers

From Kevin Grittner
Subject Re: Guidelines for GSoC student proposals / EliminateO(N^2) scaling from rw-conflict tracking in serializable transactions
Date
Msg-id CACjxUsNnf=NkeHWFD1Bo71TUfTWX7FaCghHBL3uvSsR5OZtABw@mail.gmail.com
Whole thread Raw
In response to [HACKERS] Guidelines for GSoC student proposals  (Kevin Grittner <kgrittn@gmail.com>)
Responses Re: Guidelines for GSoC student proposals / EliminateO(N^2) scaling from rw-conflict tracking in serializable transactions  ("Mengxing Liu" <liu-mx15@mails.tsinghua.edu.cn>)
List pgsql-hackers
On Wed, Mar 29, 2017 at 11:17 PM, Mengxing Liu
<liu-mx15@mails.tsinghua.edu.cn> wrote:
> Thanks, I've updated the proposal. Just one issue:
> I agree that we can make skip list a general data structure.  But
> can we use the fixed-level skip list as a Plan B? Or a quick attempt
> before the general data structure ?
> Because I am not familiar with shared memory structure and tricks
> used in it, and I cannot estimate how much time it would take.

It's not really too bad for fixed allocation shared memory, and I
can help with that.  If I thought it would save much I could see
doing a prototype without generalization, but you would still have
most of the same shared memory issues, since the structure *must*
live in shared memory.

--
Kevin Grittner



pgsql-hackers by date:

Previous
From: Alexander Korotkov
Date:
Subject: Re: Re: proposal - psql: possibility to specify sort fordescribe commands, when size is printed
Next
From: Teodor Sigaev
Date:
Subject: Re: WIP: Covering + unique indexes.