In a previous life , I designed storage / backup solutions for environments like this. Virtualised or purely physical wouldn't make a lot of difference if the whole infrastructure was based on a single, shared, storage array. The SLA in place would determine the amount of availability / redundancy needed here. For the systems described, maybe an 8-hour outage would be "affordable"... which should easily be achieved by a traditional Backup / Restore mechanism.
Of course , if the common storage failed completely , was out of support ...and no replacement could be easily sourced...then I could believe a 7-day outage, awaiting hardware for the restore ;) This is a management issue, not an IT issue.