You can collect message flow, thread, node, and terminal
statistics for message flows.
- Message flow statistics
- One record is created for each message flow in an integration
server. Each record contains the following details:
- Message flow name
- Integration server name
- Start and end times for data collection
- Type of data collected (snapshot or archive)
- Processor and elapsed time spent processing messages
- Processor and elapsed time spent waiting for input
- Number of messages processed
- Minimum, maximum, and average message sizes
- Number of threads available and maximum assigned at any time
- Number of messages committed and backed out
- Accounting origin
- Thread statistics
- One record is created for each thread assigned to the message
flow. Each record contains the following details:
- Thread number (this has no significance and is for identification
only)
- Processor and elapsed time spent processing messages
- Processor and elapsed time spent waiting for input
- Number of messages processed
- Minimum, maximum, and average message sizes
- Node statistics
- One record is created for each node in the message flow. Each
record contains the following details:
- Node name
- Node type (for example MQInput)
- Processor time spent processing messages
- Elapsed time spent processing messages
- Number of times that the node is invoked
- Number of messages processed
- Minimum, maximum, and average message sizes. When the node type
is FileInput, the message
size is reported as 0 because the size is not known at the time that
the message is propagated.
- Terminal statistics
- One record is created for each terminal on a node. Each record
contains the following details:
- Terminal name
- Terminal type (input or output)
- Number of times that a message is propagated to this terminal
For further details about specific output formats,
see the following topics: