JFS Tuning and Performance

24
qio
The qio option enables Veritas Quick I/O for Oracle database.
Dynamic file system tunables
File system performance can be impacted by a number of dynamic file system tunables. These values
can be changed online using the vxtunefs(1M) command, or they can be set when the file system is
mounted by placing the values in the /etc/vx/tunefstab file (see tunefstab(4)).
Dynamic tunables that are not mentioned below should be left at the default.
read_pref_io and read_nstream
The read_pref_io and read_nstream tunables are used to configure read ahead through the buffer/file
cache. These tunables have no impact with direct I/O or concurrent I/O. The default values are good
for most implementations. However, read_pref_io should be configured to be greater than or equal to
the most common application read size for sequential reads (excluding reads that are greater than or
equal to the discovered_direct_iosz).
Increasing the read_pref_io and read_nstream values can be helpful in some environments which can
handle the increased I/O load on the SAN environment, such as those environments with multiple
fiber channel paths to the SAN to accommodate increased I/O bandwidth. However, use caution
when increasing these values as performance can degrade if too much read ahead is done.
write_pref_io and write_nstream
The write_pref_io and write_nstream tunables are used to configure flush behind. These tunables have
no impact with direct I/O or concurrent I/O. On HP-UX 11i v3, flush behind is disabled by default,
but can be enabled with the system wide tunable fcache_fb_policy. On 11i v3, write_pref_io and
write_nstream should be set to be greater than or equal to read_pref_io and read_nstream,
respectively.
max_buf_data_size
The max_buf_data_size is only valid for HP-UX 11i v2 or earlier. This tunable does not impact direct
I/O or concurrent I/O. The only possible values are 8 (default) and 64. The default is good for most
file systems, although file systems that are accessed with only sequential I/O will perform better if
max_buf_data_size is set to 64. File systems should avoid setting max_buf_data_size to 64 if random
I/Os are performed.
Note
Only tune max_buf_data_size to 64 if all file access in a file system will be
sequential.
max_direct_iosz
This max_direct_iosz tunable controls the maximum physical I/O size that can be issued with direct
I/O or concurrent I/O. The default value is 1 MB and is good for most file systems. If a direct I/O
read or write request is greater than the max_direct_iosz, the request will be split into smaller I/O
requests with a maximum size of max_direct_iosz and issued serially, which can negatively affect
performance of the direct I/O request. Read or write requests greater than 1 MB may work better if
max_direct_iosz is configured to be larger than the largest read or write request.