Thinking of relocating your Disaster Recovery (DR) site to Azure? Read this

There are many reasons why you might be thinking about using Azure for your disaster recovery (DR). Some common ones are pay-as-you-go pricing, full scalability, eschewing costly DR sites and global coverage. Such scalability and growth potential coupled with the potential cost savings and future-proofing of your DR ensure that Azure remains an extremely enticing option when compared to other relatively expensive DR sites.

But does that mean that all is perfect in the world of Azure? There are, of course, potential hurdles to consider. For starters, you’ll have to get all of your production systems Azure-ready so that your DR system operates with a minimum of fuss. It is no mean feat to get your production system working on the Azure platform from its current working state. Bearing all of this in mind, take a deep breath, avoid the marketing hyperbole and consider carefully before taking the plunge with Azure.

What are the challenges of Azure?

Your organisation will have a uniquely configured system (servers, applications and networks). So, depending on your data centre choice, there might well be variations in the migration pathways. Event the most rigorous approach to planning and preparation does not offer any guarantee that your migration will be speedy or headache-free. Until you hit the start button you won’t know just how problematic the process might be and whether or not the benefits justify the means.

When considering Azure, a deep audit of all production infrastructure is required in order to highlight any and all changes required to make things Azure compatible. It should be noted that any existing Azure accounts might impact upon your DR project. Your production systems will need to undergo Azure adjustments along with any file adjustments. Other components may need configuring depending on your technologies and testing requirements. If your organisation uses Zerto as their replication tool then you’ll need to go through Zerto Virtual Manager (ZVM) set-up.

Any potential move to Azure has a cost implication. As of February 2017 there was a 20% increase in Azure’s pricing. More importantly perhaps, it is much harder to manage your DR budget when your DR costs are not fixed, as is the case with any pay-as-you-use service. Another cost you will have to consider is the investment in Microsoft SCVMM, whichever replication tool you choose. Additionally, you need to create a virtual machine in Azure with every DR test carried out which carries an extra cost. How do you prevent storage costs associated with an investment in Azure, from soaring?

If, from a strategic point of view, you have chosen Azure for your organisation then you should also be aware of the myriad of functionality considerations. Which replication technology is going to offer you optimum functionality? How will you manage and maintain your DR? Wow will you test your DR? What RTOs and RPOs are achievable?

It is important to understand the features and drawbacks of any replication technologies. Some will work with Amazon Web Services (AWS) as well as other private cloud environments but Azure Site Recovery (ASR) is specific to Azure. Whilst ASR works with Hyper-V, VMWare and physical servers, enabling it to replicate hybrid environments as opposed to just virtualised ones, it can be difficult from a user perspective. For example, if a user has no experience of Powershell, then using automation runbooks (in complex environments) for configuration tasks such as opening specific ports for remote desktop access or for Web access, would be extremely challenging. The necessity of configuration shows that ASR isn’t capable of immediate failover, nor is failover fully orchestrated.

ASR is known to impact VMWare performance adding unnecessary complications to your production environment. Recovery times can be extended as failback with ASR involves a migration project that can actually prevent businesses from failover. Also, DR testing with ASR can be difficult and therefore often not carried out, leading to many businesses using ASR without adequate protection against downtime.

In comparison, Zerto’s latest 5.5 release offers fully orchestrated failover and also fully orchestrates failback. This means that businesses can initiate failover a lot more easily without the migration headaches of failback. This allows businesses to now utilise their DR systems much more freely; for example, as a second set of production systems or as a highly efficient testing platform. With Zerto, business-critical systems recover faster. However, Zerto cannot replicate physical servers which means finding an alternative method of replication for physical servers in those companies that operate hybrid environments.

Larger organisations will certainly be studying the potential gains of investing in Azure - namely cost savings and huge capacity. Smaller companies can also enjoy these benefits. A wise investment of time and money would be to seek assistance from organisations experienced in getting organisations Azure-ready. This alleviates the stresses of flying solo during lengthy implementation process.