1 Comment

2011-07 – Cube and Rollup Functions

Increasingly architects and DBAs are spending more time working with aggregates.  As computational power increases, so does the clients’ desire to analyze their data in a multitude of ways.  Even though we have to ability to track every event that occurs, challenges still exist in presenting that data to users in an efficient method.

1 Comment

Comment

2011-06 - Interval Partitioning

A convenient new feature introduced in Oracle 11g is interval-based partitioning.  One simply has to create a single starting partition and specify the interval for partitions above that transition point.  As data is added to the table, Oracle will automatically create new partitions as needed.  For example, consider the following table:

Comment

Comment

2011-05 – Snapshot Too Old / Commits within a Loop

Ever run into the dreaded ORA-01555 error?  It usually happens eight hours into your ten hour ETL run.  With automatic undo management in Oracle 9i, this happens less often, but can still be a major headache.  So you look at your ETL to try to find ways to improve and discover code like this:

 

 

Comment

Comment

2011-04 – Logging

A simple, essential part of PL/SQL coding and debugging is logging. However, it seems to be one of the most overlooked parts of coding.  Below is a simple package body (the package is left out for brevity) and the DDL needed for it:

Comment