Disaster Recovery Best Options

[discoveranother]

Hi all,

 

Not sure this is the correct forum as I couldn't see a 'backup' forum, but here goes.

 

I'm looking for a solution or the best option for disaster recovery. For example, backing up a whole physical server and virtual machines that can be restored relatively quickly. Virtual machines are obviously quicker to restore but I wondered if there is a solution for backing up both physical and virtual machines to cloud and should the on-premise physical server completely crash, the customer can login to the cloud and continue using the backed up server or the backup can restore to hardware on-premise relatively quickly.

 

I know this is a lot to ask, and I know there are options like Azure to use instead etc, but the customers are on tight budgets, so just thought I'd throw this in and see what the feedback is, as some customers just have physical servers, others have physical servers with hyperv running on them with VMs.

 

Basically, i'm looking for speedier options for recovery for example, if a physical server goes down completely, even if it has VMs, that server either needs rebuilding or a new one installed and then the VMs restored or the physical DC rebuilt completely which all takes time. Is there a recommended solution, software that others use for quicker recovery ?

 

I look forward to hearing what others have implemented or recommend.

 

Thanks all

[mavhc]

What kind of disaster? Site unavailable, everyone working from home? Or just the servers all need recreating because a virus?

 

If people are working from home then go cloud, also means you can gradually move to cloud if you wish.

 

Otherwise you're just looking at restore speed, how fast would it take to redownload everything at max internet speed?

 

In theory you could download the VM on demand, but then your online storage is way more expensive I bet, given the 1 in 1000 chance you'll need to

[strawberry]

So they want the best but they want it cheap? Sounds like a facebook post. Have they tried Fablon?

 

You need to think about your services and what disaster they want to recover from and how much time the recovery should take. There isn't a one stop approach.

If premises damage is a concern then some sort of cloud storage would be good, then you'd need to provide an alternative routing option onsite.

If data loss is the concern then you'll need to buy a shed load of storage, back it backup often and do a daily tape backup as well.

If server failure is an option then you'll need a second server room with redundant hardware, set it to replicate with a delay.

If downtime is an issue then you may need to redesign your servers and replicate in real time.

If access to site is a concern then you might want to implement a remote access tech like AOVPN, RDSGW and a terminal server.

 

Or a mix of the above.