DB size

[keneso]

How big would a DB get with the following info in it?

id
name
mail
fax
tel.
in
out
position

Per let's say 20K ids?

Thank you

[CrYpTiC_MauleR]

you could you please repost it but with column types and attributes. Those all effect size.

[keneso]

Sorry, I forgot to mention I don't know anything about DBs, among other stuff  :)

I was thinking of an output page that will show the info this way

Name | name@xyz | 123-4567 | 123-7654 | 7.30am ! 8.25am | 001

Does that help?

[Kevin3NF]

Need the field types and sizes, as well as the platform.

An Access text(255) is different than a SQL Server varchar(255)
and a SQL Server varchar(255) is different from a SQL Server char(255)...

[Anonymous]

And some DBs will only create pages of rows, not just rows. A page MAY contain a LOT of rows.

If you are using a varchar column, and the row will be examined/filtered a LOT, this will take more time to process (overall) than a column that is char (fixed length). Though a fixed length column will take more space as it contains all those annoying trailing blanks.

I would use varchar on columns whose values can genuinely be variable in length. Columns which are limited in size should tend towards being fixed in length. There are no hard rules on this though, just preferences.

I read that there should be a limit to the amount of dead space used in a table. Not sure where, but probably in MS SQL BOL somewhere.

[keneso]

Thanks for the input, looks like I have to understand lots of stuff before venturing into this project.

[VGR]

naaaah ;-)

All the people above are either too rigid to provide an answer, or make fun out of you :D

You ***can*** assess grossly the size of your DB; I would personally answer quickly "this is peanuts anyway", but here we go :

I am using standard SQL row types definitions. For the varchar vs char "difference", just memorize that char(x) is what you mean by a "textual field of length x" (always x, because it's padded with whitespaces up to length x), and that "varchar(x)" is just the same except for the padding ;-))
   
So :   

id -> integer(4)
name -> char(30)
mail -> char(30)
fax -> char(15)
tel. -> char(15)
in -> time = char(8)
out -> time = char(8)
position -> char(10)

Per 20K ids, (rows) this amounts to 20k*110 octets = 2200 k octets = a bit above 2 MB

regards