Running Atlassian Data Center Products on Azure

The organization was facing a very common problem. Perhaps you can relate. They were running the Server version of Atlassian Confluence, but they were clearly reaching the limits of this configuration. Even with less than half of their employees making use of Confluence it had become unstable, and the unexpected outages had become quite disruptive.

When they started searching for a way to open up Confluence to the entire organization while also offering the availability demanded by the business, the solution became obvious: Migrate to Atlassian’s Data Center.

Why move from Server to Data Center?

Atlassian’s Confluence Server is a single-application server product, while Data Center is Atlassian’s clustered product that distributes load to multiple application servers. As the example above illustrates, the two primary reasons why you may want to move from Server to Data Center are scalability and availability:

  • Scalability – Because of the large amount of usage that the Atlassian tools tend to receive, a single server environment often just doesn’t cut it. It slows down. It crashes. It becomes a problem instead of a tool.With Atlassian’s Data Center products you can scale your environment horizontally, adding more application server nodes as your load increases. Problem solved!
  • Availability – When you start to look at the cost of having Jira, Confluence or BitBucket down for an extended period of time, you realize that clustering the environment to reduce unexpected downtime is clearly the way to go.Plus, an increasing number of IT organizations are now owning the Atlassian tool suite. They want to be able to offer a Service Level Agreement with a guaranteed uptime commitment to their customers. This simply isn’t possible without a clustered environment.

However, you do need to be aware that Atlassian Data Center only addresses application scalability and availability. It does not address scalability or availability for your database or your load balancer. When you create the overall architecture for your environment, these issues must also be addressed.

Why choose to host your Atlassian Data Center in the cloud?

What we’re seeing in the field is that organizations choose to host Atlassian Data Center applications in the cloud for the same reasons why they’re moving other critical infrastructure to the cloud Cloud service providers offer both robustness in their design and flexibility of deployment. They allow you to focus on your core business competencies without having to worry about the infrastructure side of things.

Should you host Data Center in Azure or AWS?

Atlassian currently supports two cloud service providers: Microsoft’s Azure and Amazon’s AWS. Both of these providers offer multiple availability zones in different geographies, native database and load balancer technologies, and the ability to do configuration management (i.e. manage whatever you’re running there as a configuration file instead of as a server that you log into).

Both Azure and AWS also pose unique challenges in being able to achieve the Data Center performance your team is looking for. Whichever you choose, there’s a significant effort involved with tuning the Atlassian environment to provide the expected user experience for your customers. AWS is a bit ahead of Azure in the native capabilities offered, but it’s been around a lot longer than Azure, so it is more mature.

That said, your organization might already have a relationship with Microsoft that would make Azure a very attractive option for you. Many organizations, such as the company in the example above, even have global initiatives to move as many critical business functions as possible into Microsoft’s Azure. When this is the case, hosting their Data Center in Azure is clearly the way to go.

We’re offering a webinar on “How to Run Atlassian Data Center Products on Azure”

If you’re looking for a way to make the most of your Microsoft contracts while bringing scalability and availability to your Jira, Confluence or BitBucket installations, this webinar is for you. Coyote Creek has extensive experience getting Data Center set up on Azure, and we’ve developed a methodology that works.

You will learn how to:

  • Learn about Atlassian’s deployment templates for Azure (and find out why these should only be viewed as starting points.)
  • Configure Azure to suit your needs
  • Make the most of Azure’s native functions
  • Improve performance in Confluence Data Center on Azure using the testing methodologies we’ve developed
  • Avoid the common pitfalls that can turn the migration into a nightmare

As part of this webinar we’ll share the journey we followed to architect and complete a 15,000+ user implementation of Confluence Data Center in Microsoft Azure—the good, the bad and the ugly.