JIRA MultiSite or JIRA Clustering?

Both JIRA MultiSite and JIRA Clustering add additional servers to your JIRA deployment to improve performance, scalability and reliability. JIRA MultiSite accomplishes this by bringing data closer to users at each location. JIRA Clustering accomplishes this by balancing workload across multiple clustered JIRA servers at a single location that appear to the end user as a single JIRA instance.

Your answers to the following questions will help determine which deployment option makes the most sense for your IT organization:

1. Where are your users currently located, and where will they be located in the future?
   1. Are users located across a high latency or low bandwidth wide area network (WAN) connection?
   2. Are users depending upon an unreliable WAN connection to access your current JIRA server?
      If the answer is yes to either question, consider JIRA MultiSite, which co-locates the JIRA server closer to these users.
   3. For each location, when do users typically access your current JIRA server?
      JIRA MultiSite supports a Follow-the-Sun model that allows different servers (nodes) to be scheduled for designation as the Distinguished Node. This allows performance to be optimized for each location during its normal business hours, and allows each location to access JIRA during normal business hours, even when other sites are down or unavailable.
   4. Are a large number of users concentrated at a single location?
      The workload from a large set of users at one location may still require JIRA Clustering. If your organization has a large majority of users at one location and small groups of users at remote locations, JIRA Clustering may solve your scaling and performance issues. As a browser based application, individuals located far from the server may still receive reasonable performance. Without a concentrated set of remote users, JIRA MultiSite may not be warranted.
      Conversely, if most of your users are concentrated at a single location, but there are other smaller, yet still significant groups of users at remote sites, a combination of JIRA Clustering (at the large site) with JIRA MultiSite (with single JIRA replica nodes at the remote sites) may be best.
2. Are IT application management best practices driving JIRA deployment needs?
   1. Is there a requirement for databases to be located on database servers in a central data center?
   2. Is there a requirement for application servers to be located in a central data center?
      If the answer is yes to either question, IT requirements may overrule other considerations. In this case, performance can still be enhanced by deploying JIRA Clustering. In addition, your IT organization will appreciate the fact that individual servers in the JIRA cluster can be taken off-line for periodic maintenance or upgrades without forcing JIRA to be shutdown and unavailable to end users.
3. Is there a requirement for your JIRA data to be stored off-site?
   1. Do Corporate Governance requirements necessitate that JIRA data be stored off-site?
   2. Does a Disaster Recovery plan require an off-line server for business continuity?
   3. Do government regulations require that data be stored within a region?
      If the answer is yes to any of these questions, consider JIRA MultiSite which distributes active copies of your JIRA data to other locations, and provides failover and automated disaster recovery, without requiring any third party solutions.
4. Is control of JIRA data split across multiple business entities?
   1. Are two or more corporations in a partnership?
   2. Are multiple business units that are not co-located sharing the JIRA data?
   3. Are multiple product teams or projects sharing the JIRA data?
      If the answer is yes to any of the above, consider JIRA MultiSite which will distribute active copies of your JIRA data to other locations, thus providing control of the data to each entity.