Jira Data Center Performance

Testing results summary

• Under a high load, clustered Jira has improved performance overall.
• Throughput and capacity show a near linear increase as nodes are added to the cluster.
• Specific actions show varying improvement in response times. For details, please see Testing results for specific actions.
• You might observe a different trend/behavior based on your configuration and usage. For details, please see What we tested.

Testing methodology and specifications

The following sections detail the testing environment and methodology we used in our performance tests.

How we tested

Our performance tests were all run on the same controlled isolated lab at Atlassian. For each test, the entire environment was reset and rebuilt. The testing environment included installing the following components:

• Apache load balancer 
• Postgres database and the required data
• Clustered Jira on one, two, or four machines as required for the specific test
• Chrome browser on 'x' number of machines as required to generate the load for the specific test

To run the test, we used a number of machines in the lab to generate load using scripted browsers and measuring the time taken to perform an action. An action here, means a complete user operation like adding comment. Each browser was scripted to perform an action from a predefined list of actions and immediately move on the to next action (i.e. zero think time). Please note that this resulted in each browser performing more tasks than would be possible by a real user and you should not interpret the number of browsers to be equal to the number of real world users. Each test was run for 20 minutes, after which statistics were collected.

What we tested

• All tests used the same Postgres database containing:
  • 550 projects
  • 628 custom fields
  • 450,000 issues
  • 10,000 users in total

• The mix of actions we included in the tests represented a sample of the most common user actions* representing three typical types of users: Issue Browser, Issue Creator and Project Manager. The table below show the ratio of actions performed by each of these users. These user-based actions were repeated until the test was completed.

* The tests do not include admin actions as these are assumed to be relatively infrequent. 

Hardware

All performance tests were all run on the same controlled, isolated lab at Atlassian using the hardware listed below.

Hardware testing notes:

• In order to quickly put more stress on the Jira nodes with less load, cluster nodes were set to use only 2 cores out of 6 from each CPU, thereby reducing its processing power to one-third.
• For Jira instances being tested, 9 GB of memory was allocated to the JVM consistently across all tests. This may not be optimized for all cases but allowed for consistency and comparability between the tests. 
• During the tests we did not observe high CPU or IO load on either the database or load balancer servers.
• During the tests we did not observe running out of HTTP connections in the load balancer or connections to database.
• The browser and servers are in the same location so there was very low latency between client and server.

Testing results for specific actions

The following tables show the relative performance of the various actions in the tests as the load increases for each Jira instance configuration: Jira Server (one node equivalent), two node Data Center, and four node Data Center. Each table shows the response time relative to the baseline response time which we determined to be Jira Server with six browsers.