Re: IBM i Hosting - Disaster Recovery

[Brian <belstsrv@xxxxxxxxx>]

Hi Rob,

That's definitely what we want.  We need the hosting company to be 
geographically diverse from our primary location (USA - East coast) so 
taking tapes is definitely not what we want.

My thought right now is to use something like your EMC device to 
replicate our backed up data to the remote location and then have that 
data there for restoration in the event of a disaster.  A hot swap is 
not required so I do not think QuickEDD is something we need to pursue 
(although a good product, IMO).  I think replication of nightly backups 
to the other facility with the ability to restore them fairly quickly to 
resume operations is good at this time.  I'll call that a warm site type 
of solution.

It sounds like you have hardware (another IBM i) at the other facility 
though, correct?  That's what I am trying to avoid, if possible via a 
hosting solution.

Example scenario:

IBM i hosting company allows us to keep an EMC type device at their 
facility.  We back up locally, it replicates over to them.  We 
rent/reserve (at a cost) space to allow us to come up on their hosted 
solution in a DR situation.

What I am not sure about is if we could keep a device at their location, 
cost of bandwidth to allow replication of backups to that location, cost 
of reserving DR space, etc.  There may be other ways to get our data to 
their facility too that would not involve our equipment, if they offer 
something to allow us to replicate over to them.

I'm for sure on board with your train of thought.  I just need to figure 
out some of the specifics with someone that does this sort of thing and 
get some costs/options.

Thanks to all for your feedback so far!

On 5/2/2018 8:47 AM, Rob Berendt wrote:
> In general you may want to look at something different than the "we'll
> rush tapes up to that site" kind of solution.
> With our solution we shutdown and run on the other site once a quarter. We
> do that when the remote site is unoccupied with no humans inside.
> We also do backups at that site so as not to interrupt operations on
> "production".
>
>
> Rob Berendt