Instant VM Recovery at 40,000 Feet… Streamed from Amazon S3!
Last week I wrote about a cool new enhancement coming in v10 of Backup & Replication where we are introducing a Mount function that will enable users to import backup restore points for recovery purposes in the case of disaster. Those restore points being located on an Object Storage Repository initially backing a Capacity Tier extent of a Scale Out Backup Repository. This can even be done with the Community Edition (without the need for SOBR) which means that the Cloud Tier now becomes a mechanism for recovery from any device to almost any platform.
Did you do IR from S3? 🙂
— Anton Gostev (@gostev) November 16, 2019
Challenge accepted… and a few weeks later I was able to put the new Mount Object Storage Repository enhancement to the test while also looking to find out just how well our Cloud Tier Intelligent Block Recovery technology handles the efficient streaming of Object Storage data for instant recovery. I performed an Instant VM Recovery direct from an Amazon S3 Object Storage Repository to the local ESXi server on my laptop.
I recorded a video of the proof of concept which can be viewed below:
While a little slow given the latency to the US-EAST-1 S3 Bucket, I still had the VM mounted and booting while flying at 900km/hour at 41,000 feet. That’s a pretty amazing example of how innovative Veeam’s Cloud Tier technology is and the smarts we have built in under the hood that differentiates it from other implementations. The key being that there is no pre-staging of the data required and we are aware of what objects of data we need to stream for recovery!
Not only was that a fun experiment to try, but it also shows the power of being a software driven platform that is agnostic of any hardware of filesystems. It shows true portability and recoverability of data…even in the most unlikely of places and circumstances!