Folks with on-prem workloads, what is your DR setup? Do you DRaaS?

[cerberus]

Would be interested in hearing what other folks out there are doing for DR these days.

I'll go first, we have an on-prem datacenter, fully managed in-house VMware 3-node cluster (75ish VMs, 10TB ballpark) with block storage SAN, production VMs are backed up by Veeam (backup-to-disk-to-tape). Rock solid setup that we keep current.

For our D/R setup, it is identical to production running on used equipment that is pulled from production, no issues. We use our D/R site frequently for lab work, cloning VMs and testing configuration changes/updates. The initial replication window was slow (we had to seed the replica of large multi-TB file servers), but the changed blocks nightly replication jobs are fine and finish within 6-8 hours.

This has been in place for well over a decade, and while it works, and I am curious about the advances in DRaaS solutions and other off-prem managed solutions.

Would be great if anyone has first-hand experiences to share where they lifted their DR into a DRaaS solution, did you use a DRaaS vendor or build it yourself on AWS/Azure?

[tyler.jurgens]

Really depends on what you are looking for and what gaps you are trying to fill. Going to a managed DR provider can have many benefits:
- You don't have to manage your own DR site (from hardware to datacenter services), etc.
- You can partner with a company that specializes in providing DRaaS
- Often you can chose multiple different locations for your DR site
- Plus many more.

You can build it yourself as well of course, but then you have to build and manage it yourself. That may or may not be a big deal of course.

Disclaimer: I work for a DRaaS provider.