For evaluation purposes, where there will only be a small number of users, Jira will run happily on any server that has a 1.5 GHz processor and 1 GB to 2 GB of RAM. As your Jira usage grows, a typical server will have a quad core 2 GHz+ CPU and 4 GB of RAM dedicated to the Jira application, and at least 10 GB of hard disk space for your database.

For production deployment, as in most applications, it is recommended that you run Jira on its own dedicated server. There are many factors that you should consider when deciding the extent of the resources to allocate to Jira; keep in mind how Jira will scale and grow. When deciding on your hardware needs, you should consider the following:

• The number of active (concurrent) users in the system
• The number of issues and projects in the system
• The number of configuration items, such as custom fields and workflows
• The number of concurrent users, especially during peak hours

It can be difficult at times to estimate these figures. As a reference, a server running with over 2.0 quad core CPU and 4 GB of RAM will be sufficient for most instances with around 200 active users. If you start to get into thousands of active users, you will need to have at least 8 GB of RAM allocated to Jira (JVM). Once you have gone beyond a million of issues and thousands of active users for a single Jira instance, simply adding raw system resources (vertical scaling) will start yield diminishing returns. In such cases, it is often better to consider using the data center edition of Jira, which offers better scalability by allowing you to have multiple instances clustered together (horizontal scaling), with the added benefit of providing high availability.

Officially, Jira only supports x86 hardware and 64-bit derivatives of it. When running Jira on a 64-bit system, you will be able to allocate more than 4 GB of memory to Jira, which is the limit if you are using a 32-bit system. If you are planning to deploy a large instance, it is recommended that you use a 64-bit system.