Database Reference
In-Depth Information
Total number of roundtrips (SQL*Net message from/to client): 3
CPU usage breakdown
------------------------
parse CPU: 0.00s (3 PARSE calls)
exec CPU: 1.50s (4 EXEC calls)
fetch CPU: 0.00s (1 FETCH calls)
The session level resource profile ends with detailed information on the apportionment
of CPU usage. Sessions that modify many rows have the highest CPU consumption in the “exec
CPU” category, whereas sessions that are mostly reading will have most of the CPU usage
accounted for as “fetch CPU”.
Session Level Statistics
The next report section contains statistical information. Transactions per second are calcu-
lated based on R and
XCTEND
entries. Entries of the form
XCTEND
rlbk=0,
rd_only=0
are used as
transaction end markers. The division of the number of transaction end markers encountered
by R yields transactions per second. What is a transaction? Of course, this is entirely dependent
on the application. Different applications perform different types of transactions. This figure
may only be used to compare the performance of the same application before and after tuning.
Note that the response time of a code path may be much improved after optimization, while
transactions per seconds may have dropped. This would happen after the elimination of
unnecessary commits.
Statistics:
-----------
COMMITs (read write): 1 -> transactions/sec 0.088
COMMITs (read only): 1
ROLLBACKs (read write): 0
ROLLBACKs (read only): 0
rows processed: 1
cursor hits (soft parses): 3
cursor misses (hard parses): 0
consistent gets: 2
db block gets: 8
physical reads: 1
buffer cache hit ratio: 90.00%
Physical read breakdown:
------------------------
single block: 1
multi-block: 0
Latch wait breakdown
------------------------
Enqueue wait breakdown (enqueue name, lock mode)
------------------------------------------------
