Let’s talk about how to maintain the physical elements in your Cloud Control environment. Cloud Control can scale for hundreds of users and thousands of targets, but only if you’re diligent about maintenance for the life of the deployment. Such maintenance is necessary regardless of the workload or size of your site. The cornerstone of software maintenance is periodic housekeeping.
Fortunately, due to advancements in both the Enterprise Manager and the supporting database, there are some tasks that we no longer need to address. First, the Repository database runs in a version that automatically collects statistics. No additional statistics gathering is necessary for Cloud Control. Second, in earlier versions we had to schedule downtime for the Repository in order to perform partition maintenance. This task is also not required.
Another area of interest is the volume of metric data and whether it can be scaled down. The metric data that is collected and uploaded by the Agents on the target hosts and stored in the OMS Repository tables is called “raw” data. Cloud Control aggregates collected metric data by hour and by date in order to improve query performance and help reduce the size of the Repository database. One of the jobs configured in the DBMS_SCHEDULER of the Repository database is responsible for rolling up (aggregating) this raw data. Once a day the previous day’s raw data is rolled up into a one-hour and a one-day table. These hourly and daily records will have hourly and daily metric data averages, minimums, maximums, and standard deviations computed. The hourly and daily tables along with the computed statistics are used for providing the historical metric information displayed in the Cloud Control Console.
After Cloud Control aggregates the data, the data is then considered eligible for purging. A certain period of time must pass for data to actually be purged. This period of time is called the retention time. The raw data, which has the highest insert volume, has the shortest default retention time, which is seven days. Consequently, seven days after raw data is aggregated into a one-hour record, a raw data point is eligible for purging. Hourly rollup metric data is purged after 31 days. Finally, the highest level of aggregation, the one-day rollup, is kept for 12 months.
It is possible to modify the default retention and purging policies. However, Oracle strongly discourages reducing any of these values. The Cloud Control default aggregation and purging policies were designed to provide the most available data for analysis while still providing the best performance and least disk-space requirements for the Management Repository. As a result, you should not modify these policies to improve performance or decrease your disk space demand. However, Oracle does permit increasing the amount of raw or aggregate data in the Repository for those cases where users plan on extracting or reviewing the information using external analysis tools other than Cloud Control.