From: | Tory M Blue <tmblue(at)gmail(dot)com> |
---|---|
To: | Craig James <craig_james(at)emolecules(dot)com> |
Cc: | pgsql-performance(at)postgresql(dot)org |
Subject: | Re: bigint integers up to 19 digits. |
Date: | 2010-02-04 18:51:37 |
Message-ID: | 8a547c841002041051i53d47fe4pf3a96e22bec85200@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
On Thu, Feb 4, 2010 at 10:43 AM, Craig James <craig_james(at)emolecules(dot)com> wrote:
> Tory M Blue wrote:
>>
>> I have a column that is a bigint that needs to store integers up to 19
>> digits long. For the most part this works but we sometimes have
>> numbers that are greater than 9223372036854775807.
>> ...
>> I was thinking of changing this to a real or double precision field,
>> but read in the docs that the value stored is not always the value
>> inserted...
>
> They're actually less precise than the same size of integer. Real/double
> datatypes trade more range for less precision in the same number of bytes.
>
>> My number will always be 19 digits long and always an integer.
>> I looked into the numeric data type, but the docs say that it can be slow.
>
> If it's *always* going to be 19 digits, couldn't you make it a text or char
> field? You didn't say if this is really a number. Do you do arithmetic
> with it? Sort it numerically? Or is it just a long identifier that happens
> to only used digits?
it is an identifier and is always a number and is used in grouping and
querying. I thought I would lose performance if it is text vs an
integer/double field.
Tory
From | Date | Subject | |
---|---|---|---|
Next Message | Jochen Erwied | 2010-02-04 19:01:19 | Re: bigint integers up to 19 digits. |
Previous Message | Alvaro Herrera | 2010-02-04 18:47:58 | Re: bigint integers up to 19 digits. |