I would have to give Jochem van Dieten the credit for the block factor.
I forgot all about that. d'oh!!!!!!!!
Here is some simple unofficial timings I got. I ran a page with the same query
through each datasource type. As you
will notice the oci datasource was the slowest, but that could be because it was the
first one to hit the database
even though it was the second query in the page. [I guess you could assume that the
jdbc.oracle.oci driver was able
to make a connect to the DB much faster that it beat the MXJDBC query (Just an
assumption) ;-)]. The others got the
benefit of the cached query results in the database. I could turn this off but do not
remember how at this point.
But look at the second run. The second run and other runs to the database were
consisted. The jdbc:oracle:thin
driver was always the fastest followed by the jdbc:oracle:oci driver. The slowest was
the MXJDBC driver. This was
consistant after the first run.
I did not test any stored procedures or anything more complex. I'm assuming that the
jdbc.oracle.thin driver is
going to be limited in some functionality. The approach I plan to take in devleopment
is to use the jdbc.oracle.thin
driver for simple queries, etc. and if something does not work with the
jdbc.oracle.thin then use one of the other
two. From what I gather from various docs, the jdbc.oracle.thin driver was designed
mostly as a light weight driver
for applets and may be missing some functionality. Maybe someone could verify this
for me.
###########################################################################
### Results of first run to the database.
qryMXJDBC (Datasource=archives_metcalf, Time=263ms, Records=864) in
/global/apachedg/apache/websites/www/wwwroot/archives/metcalf/reports/qryTestoci.cfm @
10:02:53.053
SELECT
LEVEL,
id,
pid,
keyword,
SYS_CONNECT_BY_PATH(keyword,'/') as path
FROM
ndx
START WITH id = 1
CONNECT BY PRIOR id = pid
ORDER BY path
qry9iOCI (Datasource=jdbc.oracle.oci, Time=494ms, Records=864) in
/global/apachedg/apache/websites/www/wwwroot/archives/metcalf/reports/qryTestoci.cfm @
10:02:53.053
SELECT
LEVEL,
id,
pid,
keyword,
SYS_CONNECT_BY_PATH(keyword,'/') as path
FROM
ndx
START WITH id = 1
CONNECT BY PRIOR id = pid
ORDER BY path
qry9iThin (Datasource=jdbc.oracle.thin, Time=432ms, Records=864) in
/global/apachedg/apache/websites/www/wwwroot/archives/metcalf/reports/qryTestoci.cfm @
10:02:54.054
SELECT
LEVEL,
id,
pid,
keyword,
SYS_CONNECT_BY_PATH(keyword,'/') as path
FROM
ndx
START WITH id = 1
CONNECT BY PRIOR id = pid
ORDER BY path
#### End of first run to the database.
###########################################################################################
#############################################################################################
#### Beginning of second run to the database
qryMXJDBC (Datasource=archives_metcalf, Time=253ms, Records=864) in
/global/apachedg/apache/websites/www/wwwroot/archives/metcalf/reports/qryTestoci.cfm @
10:15:37.037
SELECT
LEVEL,
id,
pid,
keyword,
SYS_CONNECT_BY_PATH(keyword,'/') as path
FROM
ndx
START WITH id = 1
CONNECT BY PRIOR id = pid
ORDER BY path
qry9iOCI (Datasource=jdbc.oracle.oci, Time=215ms, Records=864) in
/global/apachedg/apache/websites/www/wwwroot/archives/metcalf/reports/qryTestoci.cfm @
10:15:37.037
SELECT
LEVEL,
id,
pid,
keyword,
SYS_CONNECT_BY_PATH(keyword,'/') as path
FROM
ndx
START WITH id = 1
CONNECT BY PRIOR id = pid
ORDER BY path
qry9iThin (Datasource=jdbc.oracle.thin, Time=172ms, Records=864) in
/global/apachedg/apache/websites/www/wwwroot/archives/metcalf/reports/qryTestoci.cfm @
10:15:37.037
SELECT
LEVEL,
id,
pid,
keyword,
SYS_CONNECT_BY_PATH(keyword,'/') as path
FROM
ndx
START WITH id = 1
CONNECT BY PRIOR id = pid
ORDER BY path
#### End of the second run to the database.
#####################################################################################
Sean A Corfield wrote:
> On Thursday, Oct 17, 2002, at 10:06 US/Pacific, Troy Simpson wrote:
> > I found some more information on configuring the JDBC-OCI drivers OTN:
> >
> > http://otn.oracle.com/software/tech/java/sqlj_jdbc/htdocs/
> > jdbc_readme9201.html
>
> That's a big readme! Yup, looks like it has all you need and more.
> You're up and running happily now with the OCI drivers? Have you had a
> chance to do any more timings?
>
> BTW, good call on the blocking factor!
>
> "I can smell your brains!"
> -- Mittens the Kitten : http://www.matazone.co.uk/theotherside.html
>
>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~|
Archives: http://www.houseoffusion.com/cf_lists/index.cfm?forumid=4
Subscription: http://www.houseoffusion.com/index.cfm?sidebar=lists&body=lists/cf_talk
FAQ: http://www.thenetprofits.co.uk/coldfusion/faq
Your ad could be here. Monies from ads go to support these lists and provide more
resources for the community. http://www.fusionauthority.com/ads.cfm
