Common Misconfigured HP-UX Resources (April 2006)

The HP-UX Buffer Cache
The HP-UX buffer cache configuration can be confusing, and the HP-UX buffer cache is frequently over
or under configured. Understanding how the HP-UX buffer cache is maintained and used can help
you determine the proper configuration for your application environment.
This section addresses the following questions regarding the HP-UX 11i v1 and HP-UX 11i v2 buffer
cache:
• What is the buffer cache?
• How does a static buffer cache differ from a dynamic buffer cache?
• How does the buffer cache work?
• How much memory is required for the buffer cache and its related structures?
• What are the advantages and disadvantages of using the buffer cache?
• Can the buffer cache be bypassed?
• Are there any guidelines for configuring the buffer cache?
Note that the Unified File Cache (UFC) introduced in HP-UX 11i v3 replaces the traditional HP-UX
buffer cache for caching of file data. He HP-UX buffer cache is still used to cache metadata (non-user
file data) for non-VxFS filesystems. Therefore, this section only applies to HP-UX 11i v2 and earlier.
What is the Buffer Cache?
The buffer cache is an area of memory where pages from the secondary storage devices are stored.
The buffer cache is used to reduce access to the secondary storage devices by storing frequently
accessed pages in memory.
Once the file data is in memory, subsequent access can be performed in memory, without the need to
access the secondary storage device.
Static Buffer Cache Versus Dynamic Buffer Cache
By default, the buffer cache is dynamic so that it can expand or shrink in size over time. A dynamic
buffer cache is tuned by setting the nbuf and bufpages kernel tunable parameters to zero, and by
setting the minimum and maximum ranges as a percentage of memory, dbc_min_pct and
dbc_max_pct respectively. The default values are:
dbc_max_pct 50
7