-z
PostgreSQL TODO List
====================
-Last updated: Mon May 28 19:56:23 EDT 2007
+Last updated: Wed May 30 15:07:10 EDT 2007
The most recent version of this document can be viewed at
http://www.postgresql.org/docs/faqs.TODO.html.
For large table adjustments during VACUUM FULL, it is faster to cluster
or reindex rather than update the index. Also, index updates can bloat
the index.
-
http://archives.postgresql.org/pgsql-hackers/2007-03/msg00024.php
+ http://archives.postgresql.org/pgsql-performance/2007-05/msg00296.php
* Reduce lock time during VACUUM FULL by moving tuples with read lock,
then write lock and truncate table
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
-
z
+
PostgreSQL TODO List
-PostgreSQL TODO List
-
-Last updated: Mon May 28 19:56:23 EDT 2007
-
+Last updated: Wed May 30 15:07:10 EDT 2007
+
The most recent version of this document can be viewed at
first. There is also a developer's wiki at
Allow major upgrades without dump/reload, perhaps using pg_upgrade
Allow server log information to be output as INSERT statements
-Add ability to monitor the use of temporary sort files
-Make 64-bit version of the MONEY data type
Allow INET subnet tests using non-constants to be indexed
e.g. box @> point
Add NCHAR (as distinguished from ordinary varchar),
Automatically create rules on views so they are updateable, per SQL99
are added after the view is created.
Add CORRESPONDING BY to UNION/INTERSECT/EXCEPT
Have pg_ctl look at PGHOST in case it is a socket directory?
Add deferred trigger queue file
-Flush cached query plans when the dependent objects change or
-Invalidate prepared queries, like INSERT, when the table definition
is altered
Add UNIQUE capability to non-btree indexes
Improve commit_delay handling to reduce fsync()
Allow free-behind capability for large sequential scans, perhaps using
Improve speed with indexes
For large table adjustments during VACUUM FULL, it is faster to cluster
or reindex rather than update the index. Also, index updates can bloat
the index.
->
Reduce lock time during VACUUM FULL by moving tuples with read lock,
then write lock and truncate table
Fix priority ordering of read and write light-weight locks (Neil)
Add idle_in_transaction_timeout GUC so locks are not held for long
periods of time
Experiment with multi-threaded backend for backend creation [thread]
existing backends is one of the difficulties with a backend approach.
Eliminate need to write full pages to WAL before page modification [wal]
of indexes on TOAST tables.
Improve selectivity functions for geometric operators
Do async I/O for faster random read-ahead of data
Add use of 'const' for variables in source tree
of result sets using new statement protocol
Add pre-parsing phase that converts non-ISO syntax to supported
All backends running as threads in a single process (not wanted)