Horton
Menu

Re: DELETE vs TRUNCATE explanation

Andrew Dunstan <andrew@dunslane.net>

View thread
From: Andrew Dunstan <andrew@dunslane.net>
To: sthomas@optionshouse.com
Cc: Craig James <cjames@emolecules.com>, Tom Lane <tgl@sss.pgh.pa.us>, Daniel Farina <daniel@heroku.com>, Craig Ringer <ringerc@ringerc.id.au>, "pgsql-performance@postgresql.org" <pgsql-performance@postgresql.org>
Date: 2012-07-11T21:04:39Z
Lists: pgsql-hackers, pgsql-performance 
On 07/11/2012 04:47 PM, Shaun Thomas wrote:
> On 07/11/2012 03:18 PM, Craig James wrote:
>
>> It strikes me as a contrived case rather than a use case.  What sort of
>> app repeatedly fills and truncates a small table thousands of times ...
>> other than a test app to see whether you can do it or not?
>
> Test systems. Any company with even a medium-size QA environment will 
> have continuous integration systems that run unit tests on a trash 
> database hundreds or thousands of times through the day. Aside from 
> dropping/creating the database via template, which would be *really* 
> slow, truncate is the easiest/fastest way to reset between tests.


Why is recreating the test db from a (populated) template going to be 
slower than truncating all the tables and repopulating from an external 
source? I had a client who achieved a major improvement in speed and 
reduction in load by moving to this method of test db setup.

cheers

andrew

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Scan the buffer pool just once, not once per fork, during relation drop.