What is your Business Continuity and Disaster Recovery strategy?
Azure is so much more than just virtual machines, and the number of services in Azure is growing every month. It can easily become overwhelming with all the services, and for those who haven't started using Azure yet, it might be a good idea to start with Azure Recovery Services. This service can be combined with your existing infrastructure, and it consists of two completely separate parts, Backup and Site Recovery.
Azure Recovery Services can become your most important piece of the puzzle for your company's Business Continuity and Disaster Recovery contingency plan, popularly known as BCDR.
Azure Backup is a backup service for both what you have in Azure, but also what you have on-prem, i.e. locally. Since today's data centres are usually based on hybrid solutions, it is a clear advantage if the backup solutions are also built as hybrids with the possibility of storing the backups both locally and in the cloud.
The beauty of Azure Backup is that it is highly modular to suit both small and large businesses. The simplest building block is the backup agent, also known as MARS. To get a working backup solution up and running, all you need to do is install the MARS agent on a Windows server, physical or virtual. Then you can use Azure Backup to run backups on files and folders.
The next two building blocks are the traditional System Center DPM and Azure Backup Server. If you already have a System Center DPM, you can extend it using the Azure Backup Agent and then choose to use Azure Storage for long-term storage, for example. If you don't have a DPM already, you can install Azure Backup Server on-prem, which is actually a disguised and rebranded DPM server. You will then have access to the same features as in DPM, i.e. backup to local storage or to Azure Storage. Once you've installed either of these two building blocks, you can also run backup on virtual machines in Hyper- V, VMware and Azure as well as SQL, Sharepoint and Exchange. You will also be able to do file consistent backup of Linux machines in Hyper-V and VMware.
In general, you only pay for the storage you use in Azure, and you can also choose whether you want your backups replicated three times within the same data centre (LRS) or replicated to a secondary region (GRS). The backup data you choose to store locally is, of course, not paid for. Another nice detail is that you also don't charge for the data transfer of data to Azure when you run a backup or restore/recovery.
Azure Site Recovery
How is this service different from Azure Backup? Both back up data well? Yes, that's right. But Azure Backup secures data at a deeper level than Azure Site Recovery. You use Azure Backup to restore a specific document or file. Azure Site Recovery is used to replicate all or part of your data centre to another location in case your data centre is completely down.
A backup solution also tends to handle multiple versions, which is not interesting in a disaster recovery scenario. If your data centre burns up, you want the latest consistent, replicated version in your Disaster Recovery site. Backup solutions that need to be read back in multiple versions from tape, for example, are time consuming as you need to have the new replacement hardware installed and in place before you can even start reading back from tape.
Azure Site Recovery is thus a replication service to replicate the servers to Azure. What you can replicate are Azure, Hyper-V and VMware virtual machines as well as physical Windows and Linux servers, and it replicates to a Storage Account that can be either local or geo-redundant (LRS or GRS).
If you are running Hyper-V and System Center Virtual Machine Manager, you can also use Azure Site Recovery to replicate on-prem between two Hyper-V servers, using Azure Site Recovery as a tool to monitor and control the replication.
Azure Site Recovery automatically handles IP address changes, boot orders, manual scripts and other things that may need to be adjusted when switching over to your disaster recovery solution, using runbooks.
The beauty of this service is that you can test your disaster preparedness by booting one or more replicated servers in Azure in a fully isolated test environment, in full operation. After the test is completed, the isolated servers are automatically deleted.
You also don't have to pay for the workload you choose to replicate to Azure with Azure Site Recovery. It is only if you initiate a failover that Azure Site Recovery remaps your replicated workload to virtual machines, and which you then pay for.
Want to know more about how you can easily get started today with the backup of the future? Learn more by clicking the button below and filling in the form to download the document.