Blogger's "Stats" feature shows that the 5 most popular posts in the past 30 days have been :
I have received confirmation that I passed the ITIL v3 Foundation Examination.
Earlier today, I had a curious query that I knew was returning the wrong results.
(Fortunately, I could look at the output and say outright "this is wrong". Had I not been observant enough and not known the underlying data, I would have allowed that "wrong" query to return the wrong results and accepted them as "correct").
More than possible Oracle code and optimization bugs, the biggest reason for "wrong results" is improper referencing of objects. My query today was similar -- and it had been written by me !
The (undocumented) CARDINALITY Hint has been available since --10g-- 9i (correction by Timur). This allowed the Developer / DBA to "hint" to Oracle the expected number of rows that will be returned from a table that is part of a query.
Finally ! Oracle has introduced a "CONCURRENT" preference in DBMS_STATS routines to allow a gather stats on multiple objects concurrently.
Thus a GATHER_SCHEMA_STATS call can be made to have Table Statistics gathered on multiple tables concurrently.
In earlier versions, the DEGREE (of parallelism) would only apply to large tables. Thus, in a schema of some thousands of tables, the GATHER_SCHEMA_STATS call would run through all the tables serially. Only at some large tables would it use Parallelism.
A recent forums question was about backup and recovery to be synchronised for two databases.
To reply, I used my knowledge from the "Distributed Databases" documentation in V7 and how we used to implement Advanced Replication.