Databases Reference
In-Depth Information
In this recipe, we have seen various statistics of the on-buffer Cache. We can also obtain this
data and the Buffer Cache Hit Ratio from
STATSPACK
reports, as shown in the excerpt in the
following screenshot:
To obtain optimal performance, we can configure multiple Buffer Pools for different segment
usage needs, as we have seen with the
KEEP
and
RECYCLE
pools.
The
KEEP
pool is specifically designed for situations where we want to store some data
in the Buffer Cache because it is used very often. We can also store small tables, typically
subject to full table scans in the
KEEP
pool. We will use the
RECYCLE
pool, instead, to store
segments rarely accessed or (almost) never reused, so they don't contribute to the age out of
other segments in the Buffer Pool.
See also
F
The
Avoiding
Full
Table
Scans
recipe in
Chapter
4
,
Optimizing SQL Code
