| From: | Christophe Pettus <xof(at)thebuild(dot)com> |
|---|---|
| To: | yudhi s <learnerdatabase99(at)gmail(dot)com> |
| Cc: | pgsql-general <pgsql-general(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: How to add columns faster |
| Date: | 2024-03-03 20:00:03 |
| Message-ID: | 1C6E9170-65F0-4F16-85E1-97C422699702@thebuild.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
> On Mar 3, 2024, at 11:40, yudhi s <learnerdatabase99(at)gmail(dot)com> wrote:
> Thanks for the clarification. In case of adding the column as volatile default (like current_timestamp function as default) or say adding NOT NULL column with some conditional population of existing values will be a full table rewrite. In such scenarios, the full table rewrite operation is going to take a long time , so what will be the fastest way to achieve that with minimal to no downtime?
For other situations, you probably have to create the column as NULL-able, and then backfill it in groups of records until the whole column is populated. Remember that dropping the NULL constraint afterwards will require a full table read (although not a rewrite).
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Christophe Pettus | 2024-03-03 20:03:07 | Re: How to add columns faster |
| Previous Message | yudhi s | 2024-03-03 19:40:01 | Re: How to add columns faster |