| From: | Sean Davis <sdavis2(at)mail(dot)nih(dot)gov> |
|---|---|
| To: | andreas(at)junius(dot)info |
| Cc: | pgsql-novice(at)postgresql(dot)org |
| Subject: | Re: UUID, performance of primary keys |
| Date: | 2009-06-18 12:00:08 |
| Message-ID: | 264855a00906180500w6ea867b6v81fb024177e64101@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-novice |
On Thu, Jun 18, 2009 at 3:24 AM, Andreas Junius <andreas(at)junius(dot)info> wrote:
> Hi Postgres professionals,
>
> I'd like to use UUID for object id's (java.util.UUID). Such an object
> consists of two long values. I could now use two columns (int8) in a table
> for the primary key or one using the string representation. But I don't know
> which option is the more performant one or if there exist another more
> suitable solution. Does anybody know more about this issue?
>
The ints will be faster, I believe. However, there is a UUID data type in
postgresql. I'm not sure what its performance would be, relative to the
other two options.
Sean
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Lennin Caro | 2009-06-18 15:10:53 | Re: problem with sequence number using a trigger |
| Previous Message | Andreas Junius | 2009-06-18 07:24:43 | UUID, performance of primary keys |