Our client, a leading provider of personal and commercial financial banking card services, was seeking to improve the performance of their TPS (Transaction Processing System) Jira Project instance. This instance was managed from the Marketing Team and critical to support the overall business.
Our client's Jira instances crashed as often as 1-2 times each day and the client would make attempts at closing out issues they created or rebooted the application. Because of these crashes, this greatly affected the company as they accumulated several key workflow processes that stretched across marketing as well as company-wide. This limited their ability to collect, share, and deliver critical data internally as well as externally from clients.
The poor performance being experienced by end users within the TPS Jira Project primarily impacted:
Current Jira application timeouts and an inability to change the status of workflows (ie. A user could not progress an issue through the workflow without the application timing out. The user received a 'Cannot connect to Server' message).
End users not being able to perform their day-to-day work without interruptions.
Efficiency and availability of a revenue generating application.
Praecipio decided to help solve these issues with a two-phased approach that would allow for remediation:
Phase One: Praecipio would provide a Performance Assessment of the client’s global Jira instance. Our resources worked directly with SMEs to understand the problem, worked towards diagnosis of the root cause, and provided remediation recommendations. This first portion of the engagement was essential in setting up Phase Two's remediation effort.
Phase Two: Once recommendations were provided and an agreed path to move forward by both parties, Praecipio resources began iteratively testing solutions and made every effort to remediate.
Days into the Performance Assessment, we discovered that the client worked off of an internal board that would load all of their projects issues into once place - Which proved to be a key disruptor in memory storage and utilization.
Additionally, we found that the memory settings to their application was substantially higher than needed, and we fixed this by significantly lowering the memory. While on the project, we were able to experience a portion of the crash, which resulted in the application freezing and underperforming. In our recommendations documents, we also suggested our client an increase to the maxim concurrent connection and apache client database to better serve their instance going forward.
Praecipio was able to help our client resolve their application outages completely. This resulted in the client being able to work more effectively and efficiently within their instance. In addition, we were able to clearly validate what the internal teams were doing properly, as well as resolved immediate performance blockers. Lastly, we provided our client a tangible knowledge-base recommendations document for future best practices.