Oracle database management, tutorials, scripts, coding, programming and tips for database administrators
UNDO is a key part of any Oracle transaction, and Automatic Undo makes undo management easier for the DBA. Read on to see how it works and why most undo-related errors are a thing of the past.
If it weren't for the Hakan Factor, bitmap indexes wouldn't exist. Read on to see what the Hakan Factor is and how bitmap indexes use it.
Finding 'unused' indexes in Oracle isn't an easy task for a number of reasons. Read on to see why an index you might think isn't used probably is and discover the long path to dealing with such indexes.
Declaring a primary key isn't always easy, as some choices that look good at the start aren't the best choices when they're examined in depth. Read on to find out what data would make a good key, when to use a surrogate key and when to combine the two.
Delayed block cleanout can be frustrating when it generates errors. Read on to see what it is and what you can do about it.
The index property 'clustering factor' relies on the table data location, which is why rebuilding an index won't affect it. Read on to see what the clustering factor represents and how 'fixing' it for one index can make it worse for others created against the same table.
Over the years Oracle backup and recovery has changed significantly, from manual backups to the most recent version of RMAN. Read on to see where we were and where we are now.
Monitoring tablespace space is a constant, yet necessary, task. It's changed over the years and releases, so read on to see how it used to be done and what new views are available to make the task easier.
According to Jonathan Lewis: "Stoptimisation - the art of optimization by not doing the things you don't need to do." Read on to see what things you may be doing that you shouldn't be doing.
When using BULK COLLECT it's possible to see all of the exceptions thrown for the bulk collection by coding the exception handler properly. Read on to see how to implement such an exception handler.
NULL values in a table can cause Oracle to decide to use a Cartesian join with large data sets. Read on to see one way this can be addressed.
In Oracle 12c an interesting problem arises with ANSI joins, star expansion and tables with 500 columns or more. Read on to find out what happens and how you can work around the problem.
SGA and PGA management can be confusing. Oracle can make it easier with two views that provide advice on how to set the associated parameters. Read on to see what those views are and how to read the data.
In Oracle 11.2 using analytic functions can generate 'interesting' cardinality values that are far larger than they should be. Read on to see just how far off those values can be.
Overflow data in an index-organized table is usually put in its own tablespace. Sometimes it isn't; read on to see what overflow data is and how to change where it's located.