I think this might be related to the current address flooding happening on the network.
My observation there was:
This is likely updating the timestamps in our tables.