Jira Configuration for Amazon AWS

Question

Has anyone already configured Amazon AWS to support the following Atlassian products, rather than everyone having to calculate and estimate. Any proven figures please that would support...

Product - Product Users

Jira - 100
Confluence - 100
Fisheye - 50
Stash - 50
Bamboo - 10 (Remote Agents)
Jira Agile - 100
Jira Service Desk - 100

Many thanks,

David

Pavel Korsukov · Answer

Hi David,

For your requirements it looks like you have a relatively small load. The main bottleneck may be a Java system requirements itself. So, I would recommend to use no less than t2.medium instances on Amazon Web Services EC2. Also, I would deploy using the joined database for JIRA and Confluence.

In terms of time estimation in may be tricky as soon as you need multiple environments and you wish to configure it to be secure and performant. It may be a good option to use Deploy4Me deployment service to reduce your efforts on deployment and configuration. https://deploy4me.com/en/install/#Vendors=Atlassian

Boris Berenberg · Answer

Hey David,

I think that one thing to consider is that everyone has very different usage. Take for example two JIRA instances which we operate at Atlassian: https://jira.atlassian.com/ and https://support.atlassian.com/ Each are large instances, with large numbers of issues, and large user counts which are active throughout the day. However, activity in Support tends to be a balance of read and write operations because of the nature of the content. On the flip side, JIRA is mostly a read only instance. To ensure that each one performs well, we need to tune the environments to their specific needs.

I would start by looking at the minimum requirements for each of the products you are using, and just tallying those up. Annecdotally, I can tell you that customers with your user count tend to run fine on our per product default values. Of course, there are exceptions when people integrate scripts and other automation which results in higher than normal load.