Backup VS Replication?

[Vman]

Hi,

Can somebody explain to me exactly what the difference is between a replication and backup is? I setup my replication to replicate every 30 minutes, isn't a replication supposed to be continuous with every change?

Also I am a bit confused about how you can failover to a replication. If you are replicating across a VPN to a remote site, how do you failover since the IP subnet on the remote site will be different? Is that how replication is supposed to work where you power on the replication at the remote site and it takes over as the same VM you were replicating from? Lets say the replication was a web server, how would the outside world still be able to access the website from the replicated VM?

[Vitaliy S.]

Hello Vman,

Job type comparison can be found in our sticky F.A.Q. topic, please look it through. In addition here is a short summary that you may find useful:

Replication = low RTO.
Comes at a price of lower performance, and high storage usage (uncompressed images).
Target can only be another ESX(i) server (standby or production).

Backup = high RTO.
Provides better performance, and low storage usage (up to 10x less than replication due to compression and dedupe).
Target is any storage device, including tape (D2D2T).

Lets say the replication was a web server, how would the outside world still be able to access the website from the replicated VM?

You would need to re-assign the IP address for this server and re-configure routing between two subnets. Here is a similar topic, check it out. Please also note that our upcoming version (v6) will do VM re-IP automatically, so it becomes even easier to fail-over in case of a disaster.

Hope this helps!

[Vman]

Thanks i got the Backup VS Replication down to a science, i see the difference now. Now I am still confused about the IP Routing. What i would do is create a static route across the VPN to the disaster site for the IP of the replicated VM? What happens if the first site is completely destroyed or something crazy like that including the edge router? How would you get people to go to the replication/DS site? It would have to be with DNS right?

[Vitaliy S.]

If you have edge router destroyed how would you route traffic to the DR site at all?

Yes, you can reconfigure existing DNS servers to resolve failed VM's hostname to a new IP address on the DR site, but you may also want to consult with your network engineers regarding recommended IP routing configuration to verify your DR failover strategy.