hmhhh .... apparently IO is the bottleneck ... it "smells" like Something with the worker ... does the worker "lives" on SAN as well? I would first try to put the Worker on a DAS SSD .... then try to restore to SAN
[Virtualization Support for SME and NGOs | DLI Solutions ]
interesting (sadly). clearly IO buffers up to "Deadlock" .... hmhhh. My advise: open a ticket at VEEAM. P.S.: For me "Something" saturates the SAN link, but what. You could "play" with buffers but this a long exercise.