Orca for Oracle

Open Source Oracle Statistics Gathering and Trending

orca_oracle is a statistics gathering agent for Oracle databases.  In combination with orca, it presents graphs which provide intraday to multiyear views. orca_oracle is designed to assist working DBAs to gain a better understanding of the performance of their systems. 

orca_oracle is designed to be low impact and low maintenance. It uses the perfstat schema, collecting data generated by your existing snapshot mechanism.   Through orca's use of compact "round robin databases" the storage space required for each monitored attribute does not change over time whilst allowing up to three years history of daily averages. Database access is reduced to a minimum and viewing of graphs generated from the statistics gathered requires no more than a web browser.

orca_oracle is built on an Open Source basis and is available under the GPL, allowing improvements to be carried out freely and taking advantage of the community approach which open source encourages. It uses tools which have proved their worth in production environments and which continue to improve through the committed involvement of active user communities.

Orca for Oracle presents data in graphs, which are collected and displayed as web pages.  Graphs and web pages can be updated immediately new information is available, or batched for lower impact.

As is usual with orca, graphs are normally created for these time periods

It is possible to view a single performance statistic for all databases monitored in a group, or to show the statistic at each time interval simultaneously for one database.  Navigation from one view of the data to another quickly becomes intuitive.

As these graphs are images within a web page, it is a simple matter to copy them for a service report, or to collect month-end statistics for automated delivery.

Example Output

Sample daily graphs are shown  here.  For a full installation, the links would connect to other sets of pages though in this example only the links in the footer will  work.  Note that precise, detailed information is included to assist in interpreting the graph, including the time at which data was lasted added and exact figures for this last element along with the minimum, maximum and average of each value in the time period graphed.

Fuller information on the meaning of the statistics presented is on this page.


If you would like a copy of the scripts, or to find out more, email orca at hoopoes dot com . Alternatively, Hoopoes Limited can provide support on a commercial basis.

orca_oracle requires
Check the plentiful official documentation on installing statspack and setting up regular data collection. Don't forget that you will need to purge older data eventually!
Orca performs all the graphing and web page building functions. orca_oracle is, really, just another collection agent for Orca.
Perl DBI
In addition to the perl required to get Orca working, you'll need to install the DBI and DBD::Oracle modules.