PM Snapmirror Throttle schedule


Quick question..(i searched and couldnt find a definate answer)

(edited to add Ops Mgr 4.0.2 on windows)

When there is a throttle policy set in PM - how is it applied to multiple datasets/volumes?. ie will the throttle limit be set per individual mirror, or is PM smart enough to set a lower limit if multiple mirrors are running concurrently.

Thanks in advance

Colin G

PM Snapmirror Throttle schedule

At the start of each transfer, we determine the current throttle value for the entire job, divide by the number of relationships and throttle each transfer to that amount. This limits the total maximum througput of the entire job.

We do not re-adjust the throttle values as individual transfers complete.

PM Snapmirror Throttle schedule


How does PM handle protection jobs that overlap? (while still in the same throttle window)

If a protection job is still running when another kicks off, does it take the current transfers into account when setting the throttle for the next transfers?

PM Snapmirror Throttle schedule

If one job is running, the second one will wait for the first to complete before starting. The throttle values are re-computed using the throttle values when the second job actually starts working.