Re: row read performance
- From: "VisionSet" <spam@xxxxxxxxxxxx>
- Date: Sun, 29 Jan 2006 13:18:19 GMT
"joeNOSPAM@xxxxxxx" <joe.weinstein@xxxxxxxxx> wrote in message
news:1138465549.421322.83030@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> One of the basic tenets of DBMS application performance is to not
> move the
> data out of the DBMS unless you have to display it.
Yes, I have to display it, if it has changed ie Select includes a timestamp
check.
> What is the ratio of wanted rows to DBMS contents?
Tiny!
> type (forward only, insensitive) and statement.setFetchSize() to a
> reasonable
> setting for your purpose.
Yes, I'll do that.
> Then do one query, select the columns you need
Ah, so you believe that unasked for column does improve performance, ie 2
identical tables except that one table has a massive object column give the
same fetch performance if you never ask for that column.
> from the table where key
> IN
> (1,2,3,...) Ie: one list, which if you can sort at the client, it
> would be better.
I could do a bunch of
(pKey = x1 AND timeStamp > y1) OR
(pKey = x2 AND timeStamp > y2) OR ...
better than individual Selects - especially as I haven't even got a
connection pool - fairly toy like app at present.
there would likely be no more that 10 rows required
> What DBMS and driver are you using?
MySQL, ConnectorJ
--
Mike W
.
- Follow-Ups:
- Re: row read performance
- From: joeNOSPAM@xxxxxxx
- Re: row read performance
- References:
- row read performance
- From: VisionSet
- Re: row read performance
- From: joeNOSPAM@xxxxxxx
- row read performance
- Prev by Date: Re: readObject ClassNotFoundException ConnectorJ
- Next by Date: Re: row read performance
- Previous by thread: Re: row read performance
- Next by thread: Re: row read performance
- Index(es):