Re: VLDB Features
Simon Riggs <simon@2ndquadrant.com>
From: Simon Riggs <simon@2ndquadrant.com>
To: josh@agliodbs.com
Cc: pgsql-hackers@postgresql.org, Hannu Krosing <hannu@skype.net>
Date: 2007-12-11T23:57:23Z
Lists: pgsql-hackers
Commits
Same data as JSON:
GET /api/v1/messages/:b64id/commits
the thread's linked commits as JSON, with link sources.
API reference →
-
doc: Fix COPY ON_ERROR option syntax synopsis.
- f6f8ac8e75c9 17.0 landed
-
Disallow specifying ON_ERROR option without value.
- a6d0fa5ef840 17.0 landed
-
Rename COPY option from SAVE_ERROR_TO to ON_ERROR
- b725b7eec431 17.0 landed
-
Fix spelling in notice
- 58fbbc9d683c 17.0 landed
-
Add new COPY option SAVE_ERROR_TO
- 9e2d8701194f 17.0 landed
On Tue, 2007-12-11 at 15:31 -0800, Josh Berkus wrote: > Here's the other VLDB features we're missing: > > Parallel Query > Windowing Functions > Parallel Index Build (not sure how this works exactly, but it speeds Oracle > up considerably) > On-disk Bitmap Index (anyone game to finish GP patch?) I would call those VLDB Data Warehousing features to differentiate between that and the use of VLDBs for other purposes. I'd add Materialized View support in the planner, as well as saying its more important than parallel query, IMHO. MVs are to DW what indexes are to OLTP. It's the same as indexes vs. seqscan; you can speed up the seq scan or you can avoid it. Brute force is cool, but being smarter is even better. The reason they don't normally show up high on anybody's feature list is that the TPC benchmarks specifically disallow them, which as I once observed is very good support for them being a useful feature in practice. (Oracle originally brought out MV support as a way of improving their TPC scores at a time when Teradata was wiping the floor with parallel query implementation). -- Simon Riggs 2ndQuadrant http://www.2ndQuadrant.com