Posted by Earlier this year VMware acquired Datrium and from Datrium comes a new DRaaS offering. DR holds a special place in my heart because when I was at global enterprise companies, the annual if not quarterly DR tests were my responsibility. DR is not a sexy topic to be having in meetings and rarely do people understand the value in being able to recover the business as quickly as possible during an outage. When DR is needed and a reliable solution isn’t there is not the time to decide to take DR seriously. That is why I am excited to talk about VMware Cloud Disaster Recovery.
Other VMworld 2020 announcements:
https://digitalthoughtdisruption.com/2020/09/29/vmware-bringing-sase-to-sd-wan-vmworld-2020/
https://digitalthoughtdisruption.com/2020/09/29/vmware-vrealize-network-insight-vrni-vmworld-2020-announcements/
https://digitalthoughtdisruption.com/2020/09/29/vmware-nsx-t-advanced-threat-prevention-announcements-at-vmworld-2020/
https://digitalthoughtdisruption.com/2020/09/29/importance-of-vmware-nsx-t-in-the-modern-datacenter-vmworld-2020/
https://digitalthoughtdisruption.com/2020/09/29/vmware-cloud-on-aws-vmconaws-vmworld-2020-announcements/
https://digitalthoughtdisruption.com/2020/09/29/vmware-cloud-disaster-recovery-draas-vmworld-2020/
https://digitalthoughtdisruption.com/2020/09/29/vmware-vrealize-ai-cloud-vmworld-2020/
https://digitalthoughtdisruption.com/2020/09/29/vmware-vrealize-cloud-universal-vmworld-2020/
https://digitalthoughtdisruption.com/2020/09/29/vmware-cloud-on-dell-emc-vxrail-vmworld-2020/
Why DR matters more than ever?
Ransomware and cyber-attacks is a growing cause in DR. Ransomware is becoming a 2x cause of DR over natural disasters and human error that used to spark many DR scenarios of the past. In addition to the above mentioned impact on businesses, customers are now having to perform routine DR tests to be within compliance. DR is not just a check box anymore but becoming a necessary model within a company.
The above are some common challenges that companies are facing when it comes to having a modern DR solution. With all the complexities/unknowns/requirements for DR today, VMware is attempting to make this obtainable in a very affordable DR solution with their DRaaS offering.
Let’s unpack this slide for a minute.
In the middle is where the new SaaS based model of VMware Cloud Disaster Recovery comes into play.
Orchestration engine lives in the middle to define your run books
In the middle is where the replicated copies reside in steady state in the scale-out filesystem on top of AWS storage. This allows you to only use the capacity when needed
Left hand side this is your onprem datacenter where you data resides and on the far right is where you will be using VMConAWS as the failover capacity on vSAN.
On the right side in VMC you can have two methods
On-demand: capacity here is not provisioned until you need it
Pilot light option – pre-provision a small 3 x node foot print for faster recovery times – still smaller than if you had 100% of your workloads running on VMConAWS
How does the new DRaaS solution compare to the SRM DRaaS solution?
If you need best in breed RPO/RTO then SRM is the solution of choice. I have been a customer of both SRMaaS and VMConAWS as my DR site and it truly is a great DR tool.
If you do not need a 15 minute RPO/RTO, cannot allocate a huge part of your IT budget to DR, or looking for your first DR solution then DRaaS could be the best fit for your environment.
I decided to use this slide to discuss what happens during a time for failover or testing event.
Live Mount – take backup copies in the middle and present it as a NFS volume onto your VMConAWS ESXi hosts. Essentially, you’ll get an instant power on as soon as your capacity is available. From there you can migrate from your backup storage to your vSAN backed VMConAWS storage.
SaaS Orchestrator – here you are able to create your run books for the failover but also have the ability to create run books for your failback. This allows for the best in customizing your DR plans.
Failbacks – during the failback DRaaS compares the changed block data with the production site so only the deltas failback to your onprem following a DR event or test.
Audit reports – one last key feature is for those of you under compliance that need to prove you could fail over to a remote site during a DR event comes an audit report produced after a failover or test for your auditors.
Summary:
One of the push backs I get from people all the time is DR is expensive and we don’t need a 15 minute RPO/RTO. That is why I am glad to see VMware releasing a DR solution that is both reliable and affordable. Leveraging VMConAWS as your secondary datacenter takes a lot of operational debt off the books while giving you a fully orchestrated solution. If you need a RPO/RTO of 15minutes to an hour then I highly recommend you look at SRM. If you need a 8+ hours window then I recommend you look into a great backup solution. However, If you are around the 4 hour window then VMware DRaaS is a great solution to look at. As always, I hope y’all found this useful.