July 12, 2021
Azure Backup and Recovery: Five Best Practices for MSPs
SMB adoption of Azure is consistently increasing. Here's how to start planning BCDR for your clients' Azure assets.
As small to midsize businesses (SMBs) continue to migrate workloads to the cloud, solutions for backup and recovery on Azure are on the radar of many managed service providers (MSPs).
That’s because workloads in Azure require the same protections as those hosted on-premises—moving server operations to the cloud does not eliminate the need for a strong business continuity and disaster recovery (BCDR) strategy. MSPs are well-positioned to provide these services to SMBs, as many already provide BCDR services for on-premises server workloads.
Here are five MSP best practices for incorporating backup and recovery on Azure into your BCDR offering.
1. Determine your needs and drive efficiency
As an MSP, the more you can streamline your operations, the better equipped you’ll be to provide effective BCDR for your clients. So, when evaluating Azure BCDR solutions, look for products that meet your specific needs and simplify your processes.
For example, does the BCDR solution enable you to manage protection for both on-premises workloads and those in Azure? Does it offer management and visibility across all of your clients from a single pane of glass? Does it allow you to offload the “heavy lifting” of backup and recovery, so your team can focus on your core competencies?
2. Consider each client’s backup and restore strategy
Since you serve multiple clients, the Azure BCDR solution you choose should address a wide variety of deployment and recovery scenarios. Ideally, the solution will have the ability to protect and efficiently restore virtual machines (VMs) running on Azure. Look for solutions that can drill down from there to provide capabilities for full server restore.
Naturally, each client’s needs are unique, and their backup strategies must be determined on a case-by-case basis. With so many applications and combinations of virtual or on-prem, a flexible BCDR solution is essential.
3. Understand backup and restore time
To ensure client satisfaction, MSPs must be aware of their recovery point objectives (RPO) and recovery time objective (RTO) capabilities of the solution that they choose. Depending on the workload, parameters for each will vary.
There are a number of factors that can affect the total amount of time it takes to back up workloads in Azure. These include the amount of data being protected, how long snapshot creation takes, and the actual time required for data transfer. With some solutions, you may be placed in a recovery queue, which is another important factor. Depending on usage, this can significantly increase recovery time.
Taken together, executing a backup and/or restore can be a long-haul process. So, an Azure BCDR solution should also provide the ability to quickly recover server operations during the restore process, which leads us to the next item.
4. Have an Azure recovery plan
To ensure rapid recovery of client Azure workloads, MSPs need a true BCDR solution that enables fast recovery. This is typically achieved using a combination of snapshot and virtualization to create and mount a recovery VM directly from the backup. This capability is often referred to as “instant recovery.”
Solutions that enable this kind of recovery create a snapshot or “image” of a virtual machine (including its operating system, applications, and data) on the backup target. Some solutions can also replicate images to a secondary site for disaster recovery. This allows users to run operations from a backup instance of a virtual machine (on-premises or offsite), while primary server data and applications are being restored.
5. Determine if a third-party backup is right for you
Many organizations rely on native Azure services as part of their BCDR offerings. However, it's important to make sure that these services meet your needs for cost and redundancy. For example, let's look at Azure Backup.
Pricing is an important consideration. Microsoft charges a flat backup charge, which is based on the size of the protected instance, plus a charge for the storage that is actually used. Users pay for the number of instances that are protected with Azure Backup, including SQL servers, VMs, and applications servers. You can also choose between Locally Redundant Storage (LRS) or Geo-Redundant Storage (GRS).
Safeguarding data with Azure Backup can also invite significant cost fluctuations. You only pay for the data you use with the platform, which is cost-effective until a recovery operation is necessary—depending on the Backup Storage tier you are employing, Microsoft may charge a one-time data retrieval fee. This pricing structure can be confusing, so refer to Microsoft’s Azure Backup pricing page for the full details.
Finally, Azure Backup can’t protect data in the case of a larger Azure outage— this is a possibility that MSPs must consider for their clients’ BCDR needs.
Third-party backup can eliminate these concerns. With a third-party backup solution, MSPs will experience increased control over backup and recovery. A cloud-first, multi-cloud by design architecture can provide an extra layer of protection in case of a worldwide Azure outage, while quickly delivering full recovery of data, applications, or VMs. Depending on the vendor, this can also be done for a flat monthly fee, giving you predictable costs for better profitability.
As you can see, backup and recovery for Azure can get complex quickly. Staying on top of best practices helps MSPs to be prepared as your clients’ Azure needs expand.