There was a point in the enterprise story when the only place an organization could keep their servers was inside a dark room – often a climate-controlled basement, or a large warehouse full of servers. And for a long time, that worked – until cloud came onto the scene offering capabilities that traditional hosting methods couldn’t, such as better scalability and increased performance, at a fraction of the cost. For those reasons, the cloud became a viable deployment option for the enterprise.

Some initially scoffed at the idea of cloud computing, but now it dominates the world of technology. You can choose to use software as a service (SaaS) offerings of most of your applications, or to deploy your applications on infrastructure as a service (IaaS). Companies such as Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP), just to name a few, will host your data for you in the cloud. Either way, organizations around the globe are leveraging key cloud computing features and capabilities.

Rising to the occasion in the cloud

We know that the journey to the cloud doesn’t manifest as a one-size-fits-all solution, so we’ve developed a SaaS solution of our Atlassian application suite, as well as the ability to deploy our Data Center applications on IaaS. For those who have already decided to strategically leverage SaaS applications where possible, Atlassian recently announced a new Cloud Enterprise plan, which provides our most advanced cloud offering and is designed to meet customer needs for security, compliance, and administration at enterprise scale. But if your organization is just starting to consider a move to the cloud, or your industry or the location of your organization doesn’t allow you to use SaaS solutions, we’ve still got you covered. Data Center – our self-managed enterprise edition – has the flexibility to be deployed on IaaS. We’ve also built additional capabilities into our applications that can help give you the support and reliability you need as your organization starts its move to the cloud.

Did you know?

Around 42% of Data Center instances are deployed in the cloud.

Atlassian’s journey to the cloud

Just like any other enterprise, Atlassian’s move to the cloud was a journey – one that we spent a lot of time thinking about critically as the needs of our organization changed. While we are currently using our SaaS solution, we started our own journey to the cloud with Data Center.

Our Data Center instances were originally hosted on-prem, but our organization was growing. Hosting on-prem meant that we were dependent on the physical infrastructure that we built, but by changing to an IaaS cloud deployment, we could gain the added benefits of cloud-native features, such as better scalability and performance. And for a growing organization, both were a priority, which is why we knew that migrating to the cloud made sense for us. We quickly realized that there were major differences between hosting on-prem versus deploying on IaaS, and therefore lots to consider when making the move.

For example, using Confluence in our day-to-day business meant that we created a lot of pages (and therefore lots of data), but the way data is stored differently on-prem versus in the cloud. Normally, admins have to spend a lot of time focused on not losing their data during a migration. To solve this problem, we looked at our existing framework and wrote a script to help us.

But we also had to consider how to actually build out the Data Center infrastructure we wanted (something that every enterprise needs to consider when moving to the cloud). When we did make our move, we used AWS’s CloudFormation templates to get us there. We reviewed our existing infrastructure and made changes to the templates to fit the needs of our organization. This also gave us the chance to identify the best configurations for our applications. After working on these templates, we decided to make them available to customers who were migrating their instances.

Tip

The templates we’ve created are designed to deploy Data Center in an optimal way, but we recommend that your admins review them because you may want to change some of the default configurations to better meet the needs of your organization (as we said, it’s not one size fits all). Reviewing the templates will also give you the opportunity to examine at your current infrastructure and see if there are other changes you want to make when you deploy your Data Center applications on IaaS.

Leveraging IaaS capabilities

Moving to IaaS allows you to use features that have been built by cloud providers. To make sure you get the most out of your deployment, we’ve partnered with a number of IaaS vendors to enhance their cloud capabilities that will help your organization leverage Data Center even more effectively.

Deploying your instance

We’ve partnered with both Amazon and Microsoft to create Quick Start templates, which deploy production-ready instances of your Data Center applications. The AWS and Azure Quick Starts include all the pieces you need to deploy Data Center applications in an optimal environment. To deploy the template, just fill in a few parameters and configurations to define the type of instance you need, and you’ve got your Data Center applications in the cloud.

When we upgraded from Jira 7.4 to Jira 7.10, we used Quick Starts to create a new Jira instance from scratch and set up everything from the database to the file system automatically in a matter of minutes, compared to hours or days if we had had to set up a new stack manually.

National Bank of Canada

We currently offer AWS and Azure Quick Start templates for Jira Software, Jira Service Desk, Confluence, and Bitbucket Data Center. AWS Quick Starts are coming soon for Crowd users.

Using a content delivery network

Improved performance for all with CDN support in Data Center

As a way to support CloudFront, a content delivery network (CDN), we’ve built CloudFormation templates. As teams become more distributed, it becomes even more important that performance doesn’t lag. That gets more and more difficult as teams grow in different geographical locations. With a CDN, you can keep your teams connected around the globe while ensuring that they don’t suffer from any performance issues.

Monitoring and auditing your instance

CloudWatch provides basic monitoring of your application and database nodes and stores your logs. When you deploy Data Center using our AWS Quick Starts, CloudWatch is enabled in your environment by default. Because security is a concern for any enterprise, Data Center applications are built with additional security capabilities, including advanced auditing, designed to report on all global events that occur within your instance in the form of a comprehensive digital security record. As an additional storage solution, advanced auditing has built-in file externalization, which enables you to store your log files on CloudWatch for a longer period of time. That means that you can take advantage of features that are already available in your cloud environment.

Scaling your instance

One of the most significant capabilities you unlock with Data Center deployed on IaaS is clustering, which enables you to scale your applications horizontally.

Most IaaS cloud instances can be scaled vertically by changing the type of instance, but Data Center’s clustered deployment means you can also add additional nodes to your instance to scale the application horizontally. So, each of your application nodes runs an active instance of your Data Center applications independently of each other. If a node goes down, your team can still use their applications because traffic is directed to another node in the cluster, which makes Data Center applications built for high availability. And with traffic being redistributed to other nodes in your cluster, you don’t have to worry about unexpected downtime, which means you don’t need to worry about lost productivity or revenue.

I can just spin up a new node using our automation tools, migrate everything to the new one, then drop off the old one. It’s lovely on my side.

Matt Krohn, Systems Engineer at HubSpot

Here are some other ways horizontal scaling can benefit your team:

  • Conduct A/B testing: Add an additional node to your environment and see if you can identify ways to increase performance. Once you’ve finished testing, you can delete that node and roll out changes across the other nodes in your instance one a time.
  • Upgrade instance type: Use the cloud templates to add additional nodes to your instance with different configurations, without losing any of your existing nodes.
  • Set nodes for peak performance: Look at how your applications are being used and identify whether there are high performance times. With that information in hand, set a schedule that specifies the minimum number of nodes you’ll need during a specific time.

As you’re looking at your existing framework, you may decide to choose a different database. For added high availability, our Data Center applications support Amazon Aurora. Amazon Aurora provides an additional layer of resilience across your full application stack, which any type of enterprise can benefit from. This database leverages a distributed database cluster for higher fault tolerance, which means you and your teams can rest assured your applications will run reliably.

Deploying your applications on IaaS gives you a flexible infrastructure that works for your organization, whether it’s using cloud-native features, utilizing Data Center capabilities and features, or embracing new technologies, such as Docker container images, in your CI/CD pipeline.

Additional considerations

Once you’ve moved your applications to the cloud, the work isn’t quite over yet. Our Atlassian Data Center experts agree that once you’ve deployed, it’s important to set aside time to test your new environment, migrate your teams slowly, and continue to learn about Data Center and different cloud capabilities.

Your organization didn’t decide to move to the cloud in a day. Take the time to test your implementation to make sure that you’ve built the best infrastructure for your teams and your enterprise growth.

As with testing, take the time to move people over slowly. Continue to run your existing infrastructure in parallel with your new cloud deployment as you make the shift. This will give you the opportunity to catch anything well before everyone is using the applications.

And probably the most important thing you can do after you deploy in the cloud is continue to learn. Every cloud provider has its own types of documentation available to familiarize yourself with their solutions. We’ve also developed training and certifications for Data Center that will help you get the most from your new deployment.

If you’re a visual learner, check out these videos on how to deploy Data Center on AWS.

Need more?

If you are seriously considering deploying Data Center on IaaS, we have more resources to help. First, if you haven’t considered Atlassian’s product line for your mission-critical instances of Jira Software, Jira Service Desk Confluence, Bitbucket, or Crowd, our Enterprise Resources page may help you.

Want to learn about Atlassian’s journey as they moved to AWS with Atlassian Data Center products? Join our webinar to hear from Atlassian’s team of internal experts about how they tackled the migration and the benefits they’ve been able to unlock

Data Center and the cloud: what you should know...