Quickling wrote:
>
> Hi,
>
> Question 1:
> ---------------
> We've got a server app that does a lot of 'small' database reads and
> writes. We were originally using MS Access via DAO (Jet Engine) and we
> wanted to tighten up DB performance, so we've written a general ODBC
> database wrapper object, but mainly just to connect to MySQL. I figured
> there'd be ODBC overhead, but its a lot worse than I imagined.
>
> I want to know: does it make sense that our original system, connecting via
> 'Jet-engine' to Access, is actually much faster than connecting to MySQL
> via ODBC? This seems to be what's happened.
>
> I'm wondering if it has to do with the overhead of connecting to a
> server-based database via a tcp socket (even on localhost) rather than the
> direct-to-disk Jet engine; maybe because we do so many small reads/updates
> it's actually faster with Access? Any thoughts? Is it worth my time to
> look into using MySQL directly instead of thru ODBC?
>
> I'm obviously working on Windows (NT), connecting at ODBC version 2.0 to
> MySQL server 3.23, using a database converted directly from Access to MySQL
> using the cool (but unstable) DBTools GUI, which kindly retained all keys
> and indexes (which have been reviewed for speed).
>
If your tables are very small. I've found the same thing that you
did, Access was actually faster on some small queries.. When your tables
get bigger (and it didin't take much 100,000+ rows) then Access started
to go down-hill quickly, occasional corruption, unable to repair, and very
slow on many queries.
> Question 2:
> ---------------
> Is there any way in MySQL to compute, inline SQL, the difference between
> two datetime values? I couldn't find a function which could do anything
> but subtract intervals from datetimes... what I need is the interval!
>
> Many thanks in advance,
>
> Jesse
>
UNIX_TIMESTAMP(DATE2) - UNIX_TIMESTAMP(DATE1) = seconds
TO_DAYS(DATE2) - TO_DAYS(Date1) = days
---------------------------------------------------------------------
Before posting, please check:
http://www.mysql.com/manual.php (the manual)
http://lists.mysql.com/ (the list archive)
To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
