This documentation relates to an earlier version of Confluence.
View this page in the current documentation or visit the current documentation home.

Skip to end of metadata
Go to start of metadata


Confluence provides statistics about its internal caches that allow you to track the size and hit ratio of each cache and tune it for better performance (if necessary). See Performance Tuning for more information.

Configurable Caches

System administrators can change the sizes of Confluence's internal caches through the Administration Console and these changes will take effect without the need to first shut down and then restart Confluence. The maximum number of units for any of the defined cache regions can be adjusted individually.

Note that larger cache sizes will require more memory at runtime, so you should review the memory allocation of the Confluence Java process and the physical memory available on your server.

Viewing Cache Statistics and Modifying Cache Sizes

To view the cache statistics:

  1. Go to the Confluence 'Administration Console'. To do this:

    • Open the 'Browse' menu and select 'Confluence Admin'. The 'Administrator Access' login screen will be displayed.
    • Enter your password and click 'Confirm'. You will be temporarily logged into a secure session to access the 'Administration Console'.
  2. Click 'Cache Statistics' in the left-hand panel. There you will find a list of all objects cached within Confluence.
  3. Click the 'Advanced' tab for more detail. Below is an example for one of the most frequently used caches, the 'Content Object' cache.

Name

Percent Used

Effectiveness

Objects / Size

Hit / Miss / Expiry

Adjust Size

Flush

Content Object

80%

73%

4023 / 5000

374550 / 140460 / 55044

Adjust Size

Flush

About the generated numbers:

Percent Used:

=(Objects)/(Size)

Effectiveness:

=(Hits)/(Hits + Misses)

Objects / Size:

The number of entries in the cache / the number of total possible entries allowed (configurable).

Hit / Miss / Expiry:

The number of reads accessing cache where required content was found / the number of reads accessing cache where required content was not found / the number of objects evicted from the cache.

Adjust Size

Use this option to specify a different maximum cache size. Enter a new cache size and click the 'Adjust Size' button to set it.

Flush:

Flushes the cache.

For instance, to calculate Percent Used:

Percent Used = Objects / Size

Percent Used = 4023/5000 = 80%

To calculate Effectiveness:

Effectiveness = (Hits)/(Hits + Misses)

Effectiveness = 374550 / (374550 + 140460) = 73%
Icon

The clustered versions of Confluence use distributed cache called Tangosol Coherence.

Watching the Cache Contents

To see the specific items in the caches, view the cache statistics at <baseUrl>/admin/cachecontents.jsp.

Additional Notes about Configurable Caches

Changes to cache size configurations persist across confluence restarts as they are saved in the <confluence-home>/config/confluence-coherence-cache-config.xml file (or <confluence-home>/config/confluence-coherence-cache-config-clustered.xml for a clustered instance). In most cases, a Confluence administrator will never need to know about these files. However, if it is necessary to tune cache options other than the maximum cache size, this can be done by manually editing these files. See Cache Performance Tuning for details.

Important note about clustered Confluence installations

Icon

The cache configuration file is stored in a home directory of each cluster node. When a Confluence administrator changes a cache size, all running cluster nodes will automatically update their own configuration files in their respective home directories. However, if a cluster node is not running when an administrator adjusts a cache size, the /config/confluence-coherence-cache-config-clustered.xml file in its home directory will not be updated. Since cluster caches are configured by the first node to start, if a node with an outdated cache configuration is the first to start up, the whole cluster would end up using the configuration of that node. However, copying this file from one node to another would resolve this issue.

Performance Tuning

If you need to tune your application when under high usage, you may like to review this document for suggestions.

RELATED TOPICS