There is no metric logged indicating how long an update takes to replicate to a given DSA. If running synchronous multi-write replication between two DSAs, the DSA log files can be used to determine how long it has taken between when an update is received by the DSA, applied locally, replicated, and a response sent back to the client. I'm not sure if this would be enough information?
The sequence is:
1) When an update is received by the DSA the update-log and query-log will display the update request and when the update was received. For the time-log an operation timer is started.
2) The update is applied locally
3) If successful, the update is then replicated to any multi-write peer DSAs
4) Once a confirm has been received from each multi-write peer a second line is written to the update-log and query-log indicating the time of success. This can be matched against the line written in (1) using the "{connection id}.{message id}" label after the timestamp to determine the time taken to complete the update (including replication). A single line is written to the time-log which indicates the duration of the update + replication.