From: | Bryan Montgomery <monty(at)english(dot)net> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Searing array fields - or should I redesign? |
Date: | 2010-12-15 15:13:46 |
Message-ID: | AANLkTikGU7aSn-eD+7u_LLwTLt7Qq_bSfA1r9Z1iQtih@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Hello,
I have a process that is logging data from vehicles. I'm looking for
thoughts on the pros and cons of different approaches to storing this data -
and retrieving it. Different vehicles report different types of data. The
current process stores the data in a multi-dimensional array.
eg, insert into logtable values ( 'vehicle123', now(),
{{'voltage','13'},{'rpm','600'}};
However, I am not sure how I can write a query - for example to read all
records where the voltage field is less than 13. Performance in this case is
not a real significant issue.
Would I be better off redesigning and having a master / detail kind of
structure? Where the master table would have the vehicle id, timestamp and a
key to the detail table. The detail table would then have the key, the type
of measurement and then the value.
I guess a third approach would be to just have the detail table with
duplication on the vehicle id and time - for each data type recorded.
Thanks,
Bryan.
From | Date | Subject | |
---|---|---|---|
Next Message | Andrus Moor | 2010-12-15 15:15:02 | Re: How to restore from backup to 8.4.3 server using 9.0 dump/restore |
Previous Message | Adrian Klaver | 2010-12-15 14:37:42 | Re: create language 'plpythonu' on win failed |