Re: [HACKERS] Pluggable storage

From: Haribabu Kommi <kommi(dot)haribabu(at)gmail(dot)com>
To: Michael Paquier <michael(at)paquier(dot)xyz>
Cc: Alexander Korotkov <a(dot)korotkov(at)postgrespro(dot)ru>, Robert Haas <robertmhaas(at)gmail(dot)com>, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>, Thomas Munro <thomas(dot)munro(at)enterprisedb(dot)com>, Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: [HACKERS] Pluggable storage
Date: 2018-06-13 20:20:51
Message-ID: CAJrrPGcN5A4jH0PJ-s=6k3+SLA4pozC4HHRdmvU1ZBuA20TE-A@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox
Thread:
Lists: pgsql-hackers

On Fri, Apr 20, 2018 at 4:44 PM Haribabu Kommi <kommi(dot)haribabu(at)gmail(dot)com>
wrote:

>
> Apart from rebase, I have added the support for external relation to be
> stored in the
> pg_class. These are additional relations that may be used by the
> extensions. Currently
> these relations cannot be queried from SQL statements and also these
> relations cannot
> be dumped using pg_dump. Yet to check and confirm the pg_upgrade of these
> relations.
>

Here I attached rebased patchset to the latest HEAD. Apart from rebase, I
try to fix the JIT
support with pluggable storage, but it doesn't work yet.

Thanks Alexander for conducting the pluggable table access method
discussion in unconference
session at PGCon, I was not able to attend. From one of my colleague who
attended the session
told me that, there was a major discussion around TOAST and VACUUM features
support. I would
like to share the state of those two features in the current patch set.

TOAST:
I already moved some part of the toast capabilities to the storage, still
there exists majority of the
part that needs to be handled. The decision of TOAST should be part of the
access method that
handles the data. I will work on it further to provide better API.

VACUUM:
Not much changes are done in this apart moving the Vacuum visibility
functions as part of the
storage. But idea for the Vacuum was with each access method can define how
it should perform.

Regards,
Haribabu Kommi
Fujitsu Australia

Attachment Content-Type Size
0012-Table-AM-shared-memory-API.patch application/octet-stream 6.5 KB
0008-Remove-HeapScanDesc-usage-outside-heap.patch application/octet-stream 98.0 KB
0009-BulkInsertState-is-added-into-table-AM.patch application/octet-stream 10.7 KB
0010-table-rewrite-functionality.patch application/octet-stream 11.1 KB
0011-Improve-tuple-locking-interface.patch application/octet-stream 49.1 KB
0013-Using-access-method-syntax-addition-to-create-table.patch application/octet-stream 18.5 KB
0014-ExecARUpdateTriggers-is-updated-to-accept-slot-inste.patch application/octet-stream 6.5 KB
0015-External-relation-infrastructure.patch application/octet-stream 8.7 KB
0001-Change-Create-Access-method-to-include-table-handler.patch application/octet-stream 8.6 KB
0002-Table-AM-API-init-functions.patch application/octet-stream 9.9 KB
0003-Adding-tableam-hanlder-to-relation-structure.patch application/octet-stream 7.2 KB
0004-Adding-tuple-visibility-functions-to-table-AM.patch application/octet-stream 53.4 KB
0005-slot-hooks-are-added-to-table-AM.patch application/octet-stream 72.5 KB
0006-Tuple-Insert-API-is-added-to-table-AM.patch application/octet-stream 93.7 KB
0007-Scan-functions-are-added-to-table-AM.patch application/octet-stream 105.1 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bossart, Nathan 2018-06-13 20:29:12 Add SKIP LOCKED to VACUUM and ANALYZE
Previous Message Alvaro Herrera 2018-06-13 19:59:24 Re: [HACKERS] WIP Patch: Pgbench Serialization and deadlock errors