Testing CI Performance for SNMP Connections
Sending a specified number of predefined (pattern) messages to the integrated console using a message per second rate of your choice is the basic logic to determine the overall CI message throughput and performance of a SA z/OS SNMP connection to a SE or HMC.
Once the messages arrive at the ProcOps FP system, they are written to the NetView log. You can determine if OS message events are lost by controlling the message sequence numbers.
In the example shown in Figure 1, the ISQ999I message sequence number is 00004. The test case was started for a total of 00010 messages. In the ProcOps FP Netlog you should find all messages from 00001 to 00010. If one or more messages are missing, this indicates that message events were lost on the connection.
1 2 3 4 5 6
----+----0----+----0----+----0----+----0----+----0----+----0----+
+ISQ999I 12:24:01 Test Message 00004 of 00010 *** 1234567890$%&/(
7 8 9 10 11 12
----0----+----0----+----0----+----0----+----0----+----0
)=? qwertzuiop_QWERTZUIOP* _ProcOps-SYSCONS_ asdfgh+120Two REXX program utilities, ISQWTO3 and ISQTSND3 are delivered with the SA z/OS sample library SINGSAMP as members INGEI005 and INGEI006.
Both programs require specifying the total number of messages to be produced on the integrated console (CI) per call. The second parameter can be used to specify the message per second rate that the utility should try to achieve. For installation and usage information refer to the utility source members in the SINGSAMP library.
ISQWTO3 is the utility implementation for NetView environments; ISQTSND3 is a TSO implementation, if a NetView/SA z/OS environment is not available on the z/OS system to be tested.
Run the programs with different combinations of total message numbers and message per second rates. This allows you to emulate different CI message load situations.