Haut de page

Why it might be time to switch to Atlassian Data Center


Posted by
Sultan Maiyaiki

June 12, 2018

Adopting an instance of Jira, Confluence, Jira Service Desk, Bitbucket, Hipchat or Crowd may have started out innocently enough; it might have been just one team or just one instance. Yet similar to a zombie virus, the adoption of Atlassian applications are highly contagious. Spreading through the air (also know as “word of mouth”), team after team becomes seduced by the Atlassian suite’s ability to drive efficiency for software development, IT service management, project management and more.

Yet as your business scales and your Atlassian instances become essential for your team to get work done, how can you ensure that your mission-critical applications are not hindered by performance issues? Because if your Atlassian applications are down, you may as well be facing a World War Z scenario at your office door. There’s no need to hide in a bunker – with Atlassian’s Data Center deployment option, you’ll be able to scale up and provide consistent performance for enterprise teamwork.

How does this work? Data Center for Atlassian applications are a unique offer that allows you to run multiple nodes for your instance in an active-active cluster with a load balancer. This facilitates high availability (HA) and scalability of your Atlassian application by distributing incoming user requests across the nodes, thus reducing the response time and stress on a single node.

We imagine you’re reading this blog now if the performance of your Atlassian applications might not be cutting it on one node. Here are a couple of indicators that it might be time to scale to Atlassian Data Center products.

Atlassian Data Center benefit 1: Uptime

If you want to minimize planned or unplanned downtime, switching to and Atlassian Data Center product is a good option. Whether the downtime is caused by application issues such as JVM errors, database connection bottlenecks, or server disruptions (such as server resources like CPU, RAM or storage being overwhelmed), Atlassian Data Center products ensure that users are re-routed to healthy nodes and as such, users that rely on Atlassian instances to perform their business will not be affected by the downtime.

Atlassian Data Center benefit 2: Zero downtime upgrade

In continuation to the fact that Data Center ensures no downtime, another important benefit is the ability to run upgrade without outages. Upgrades are inevitable for Atlassian application; they could be due to security advisories, instances approaching end of life or even new features introduced in later versions. With Atlassian Data Center products, upgrades should no longer be nightmare where they have to be performed over nights or weekend.

The way it works is each node will be taken out from the cluster, one after the other, and upgraded. Once each of the nodes has been upgraded in isolation, the shared database will be upgraded. The nice thing about zero down time upgrades is that the whole activity can be done without noticeable interruption of service. We explained this in more detail in this article: Zero downtime upgrades for Jira Software.

Atlassian Data Center benefit 3: Exponential increase in concurrent users

Is your organization distributed across different time zones? Are there times, when multiple teams are online, your instances begins to lag? If the answer is yes, it implies the number of concurrent users are beginning to grow beyond the point where a single server can handle the load. Generally, large number of users accessing an instance at same time can lead to slower response times. In an active-active cluster setup, each node increases the capability to be able to handle concurrent users because the workloads are balanced.

Our experience with enterprises is that most of the time, the peak times are at points when one or two global teams have overlapping times when they need to work, causing high concurrent usage for your instance.

Atlassian Data Center benefit 4: High rate of data creation

Across the Atlassian suite, scalability of your business insinuates that more data will need to be created. This includes data such as issue rate in Jira, pages in Confluence and repositories in Bitbucket.

Let’s look at an example; for Jira Software and Jira Service Desk, it is generally easy to setup and customize them to match the way your team works – fields, workflows, screens and permissions are all customizable. Jira also keeps track of any changes applied to stored information. This generates a lot of data to store and raises challenges when it comes to retrieving information in real time, with every new issue adding to the performance challenge. Not to mention that new features shipped with every new version increase the actions performed by the Jira server.  With Jira Software and Jira Service Desk Data Center, each of the nodes has its own local index. Now with the users activity distributed evenly, the task of retrieving information will be restricted to the node which they are accessing rather than going to a centralized index directory.

Atlassian Data Center benefit 5: Performance degradation

Maybe you’ve passed all the health checks and are running on a supported architecture. But if the system load is heavy on a single node, an upgrade to an Atlassian Data Center product will ensure the load is evenly distributed to the nodes in your cluster. This safeguards against the effects of overloading a single node.

Overloading of server can be due to many factors, like concurrent usage of the application, automated tasks, reports and API calls. In Data Center, we are able to define via our balancing algorithm a specific node for handling certain tasks, like API calls. This is a huge benefit for enterprises to ensure services are never interrupted due to high requests.

Atlassian Data Center benefit 6: Project archiving in Jira

As we have mentioned earlier, the number of issues has significant performance implication in Jira. Sometimes, enterprises have old projects and associated issues that are no longer relevant for their current needs. But Jira Server doesn’t have a mechanism to easily clean up the irrelevant issues/projects.  Jira Data Center now has a built-in mechanism to archive projects and their issues. The native project archiving feature will remove old issues from the indexes and hide the projects from the project selector when creating issues. This makes Jira more user friendly, frees space and resources on the server and most importantly, improving the performance of Jira.

Atlassian Data Center benefit 7: Smart mirroring in Bitbucket

For development teams located across the globe, having many users access the same repository in just one location can quickly become a performance nightmare. Teams may need to wait hours to clone large repositories from the primary instance, thus significantly slowing down the development process. Bitbucket Data Center’s smart mirroring allows all repositories to stay in sync with the primary Bitbucket Data Center instance by automatically copying the repositories to the needed remote locations. More details on the advantages of smart mirroring and Bitbucket as an enterprise tool can be found here.

Make the switch to Atlassian Data Center with Valiantys!

If you’re currently plotting your survival strategy to head out the front door past the mob of users who can’t use their Atlassian applications properly, you’ve waited too long to scale up to an enterprise solution. Apparently you still have access to the Internet, so click below and check out the joint whitepaper published by Atlassian and Valiantys “Performance at Scale: 10 million issues and beyond with Jira Data Center” which defines the key indicators for switching to Data Center. You can also leave a comment below the line, and one of our Atlassian certified consultants will come to your rescue and ensure a smooth deployment for your Atlassian Data Center product.