A performance test team was testing the capacity of concurrent users of a web application
deployed on a multi-processor server. While testing 250 concurrent users, the administrator
discovers connections to the database are failing.
In response, a system administrator increased the connection pool to the maximum limit which the
database allows and increased the prepared statement cache size. The test passed, but the
application response time was slower.
How should the system administrator tune the data source properties to improve the application
response time?
A.
Change connection pools purge policy.
B.
Decrease connection pool reap time.
C.
Decrease number of free pool partitions.
D.
Decrease the prepared statement cache size.
B
0
0
Even though “The Reap Time interval also affects performance. Smaller intervals mean that the pool maintenance thread runs more often and degrades performance.” in that case the connection to db is the bottleneck and not CPU. The Reap Time interval affects the accuracy of the Unused Timeout and Aged Timeout settings. The smaller the interval, the greater the accuracy. Default is 180 seconds (3min) which of course will lead to performance degradation. In worst case this means that unused connections wont be returned to the pool 1800 seconds +180 seconds.
So B does make sense.
0
0
B is not correct:
The Reap Time interval also affects performance. Smaller intervals mean that the pool maintenance thread runs more often and degrades performance.
see
http://www.ibm.com/support/knowledgecenter/SSAW57_8.5.5/com.ibm.websphere.nd.doc/ae/udat_conpoolset.html?lang=en
It should be A the correct answer.
0
0
B
0
0
A
https://www-01.ibm.com/support/knowledgecenter/ssw_i5_54/rzamy/50/admin/help/udat_conpoolset.html
0
0
It says
Specifies the number of partitions that are created in each of the shared pools.
Partition support is always enabled. The default values of 0 should be used to enable the connection pool to pick the best values for performance. In some cases where large multiprocessor systems are used, adjusting the partition support properties might help performance.
0
0
It’s A
0
0
AC, have you passed this exam?
0
0
B
0
0
I agree that B is correct. IBM says ‘Purge policy – Specifies how to purge connections when a stale connection or fatal connection error is detected.’ But the context of question said, “when it increased the connection pool maximum limit and cache size, the test passed’. It means, the bottleneck was not the connection failure ( connections were waiting in the queue, not failing because of statement error ). I think decrease the reaps time does make sense to improve the responding time.
0
0
D
http://www.ibm.com/support/knowledgecenter/en/SSAW57_8.5.5/com.ibm.websphere.nd.doc/ae/udat_jdbcdatasorprops.html
Note: The higher the statement cache, the more system resources are delayed. Therefore, if you set the number too high, you might lack resources because your system cannot open multiple prepared statements.
0
0
I think ‘D’ is closest one to the correct answer ,if we read the question carefully
he mentioned that the admin ” increased the prepared statement cache size”
which might cause slowness and if it is decreased it will tune the performance as mentioned in below link:
http://www-01.ibm.com/support/docview.wss?uid=swg21108496
0
0
why not Answer C as here we are talking about performance not the connection issue and as per IBM adjusting “number of free pool partitions” will reduce performance issue
Condition Apply –> Above setting only work when system with multi processor. Please check below link.
https://www.ibm.com/support/knowledgecenter/en/SSAW57_8.5.5/com.ibm.websphere.nd.multiplatform.doc/ae/udat_conpooladv.html
0
0