Thread: Major Feature Interactions

Major Feature Interactions

From
"Simon Riggs"
Date:
There's a number of major features all coming together over the next
weeks, so I wanted to raise some awareness of possible areas of
interaction. My concern is more than just "will multiple patches apply
together?", but extends into whether the features interact with each
other with synergy, ignore each other or have negative impact.

The projects that concern me are the following internals projects

Plan Invalidation
Autovacuum changes
Bitmap indexes
GIT indexes
Dead Space Map (DSM)
HOT Updating
Updateable cursors
Restartable VACUUM

The potential interactions as I understand them are these: I'm not
saying this list is either complete or accurate; checking whether it is
or not is exactly the reason for this post.

Anyway, just trying to raise awareness to look for potential issues.

UpdCursors
----------Syntax changes only, no impact on other projects mentioned.

HOT
---
GITUsed to be some conflicts at patch level, now gone?Positive impact on GIT performance
PlanInvalheap_check_hotupdate() test can performed in planner, sowill no longer be a heap function
DSMHOT updates need not be logged, since they will be clearedby retail VACUUMs
BitmapIdxPotential issue with chilling tuples during CREATE INDEX
AutoVacHOT will change how UPDATEs are reported to pg_stat, byadding additional fields.
UpdCursorsShould tidscan follow chain to correct tuple? That seems tochange current behaviour even if it might be
useful.
Plan Invalidation
-----------------No known changes, other than the integrations already mentioned

Autovacuum changes
------------------
DSMIs implementing new forms of VACUUM: VACUUM and VACUUM ALL.Does reporting of row stats change? 

Bitmap indexes
--------------
GITDo we need both? I think yes, but need to see some info onwhen crossover occurs between two index methods.

Restartable VACUUM
------------------
DSMInteractions with DSM?
HOTMust use full cycles only

--  Simon Riggs              EnterpriseDB   http://www.enterprisedb.com




Re: Major Feature Interactions

From
"Merlin Moncure"
Date:
On 2/28/07, Simon Riggs <simon@2ndquadrant.com> wrote:
> There's a number of major features all coming together over the next
> weeks, so I wanted to raise some awareness of possible areas of
> interaction. My concern is more than just "will multiple patches apply
> together?", but extends into whether the features interact with each
> other with synergy, ignore each other or have negative impact.
>
> The projects that concern me are the following internals projects
>
> Plan Invalidation
> Autovacuum changes
> Bitmap indexes
> GIT indexes
> Dead Space Map (DSM)
> HOT Updating
> Updateable cursors
> Restartable VACUUM

What about the interaction of plan invalidation and the security
problems with 'security definer'?  I am especially curious if pl
functions will no longer regenerate their plans with each session for
8.3.  Is that going to be the case?  It wasn't completely clear from
the related discussions AFAICT.

merlin


Re: Major Feature Interactions

From
Tom Lane
Date:
"Merlin Moncure" <mmoncure@gmail.com> writes:
> What about the interaction of plan invalidation and the security
> problems with 'security definer'?  I am especially curious if pl
> functions will no longer regenerate their plans with each session for
> 8.3.  Is that going to be the case?  It wasn't completely clear from
> the related discussions AFAICT.

No, there's no intention of caching plans across sessions.

I'm intending to build some infrastructure in namespace.c to support
a stack of "override" search_path values, and then tweak plancache.c
to override search_path during a replan.  (It has to be a stack because
nested replan events are entirely possible.)  I think this
infrastructure will also work for attaching search paths to functions;
but we still lack an agreed design spec for how that should look to the
user.
        regards, tom lane