Re: [Sqlalchemy-users] SQLAlchemy ready for production?

[Michael Bayer]

On Jul 1, 2006, at 12:10 PM, Jonathan Ellis wrote: On 7/1/06, Charles Duffy <[EMAIL PROTECTED]> wrote: Additionally, there is a rather massive performance impact on account of this same decision -- I've seen Oracle logs indicating that 50% of the time spent by the database was soft parse time, which would have been one-time-only per query (as opposed to per instance of a given query) otherwise You're confusing bind variables with prepared statements. If your queries run so fast that parsing is really your bottleneck, then either you are working on a toy problem or you have so much money to throw at fast disks that hiring Michael to write a prepared statement cache for SQLAlchemy would not be a problem :) bind parameters allow an enormous performance increase on Oracle.  its not just the parsing of the query, its the plan creation that takes a lot of resources.  a complicated oracle query can take 10-15 times longer to execute the first time then on subsequent executions when its pulled from oracles cache of planned-queries, i used to see that all the time at MLB (and the dbas, who dealt with oracle crashing on a weekly basis, grepped every query made for literal params and made the corresponding dev team change every single one to a bind param).

Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642

_______________________________________________
Sqlalchemy-users mailing list
Sqlalchemy-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sqlalchemy-users