we want to have a reporting about the development of remaining estimates and hours spent on an Epic level. We’re using the standard JIRA hierarchy down to the Subtask-Level, but issues are distributed across several projects.
I have already tried to do such a report with help from your customer support, but unfortunately ran into difficulties, such as
eazyBI cannot handle reorganization of epics (Stories are moved between Epics leads to the original epic losing this remaining effort historically)
The sheer amount of data takes long to process and leads to Out of memory or Timeout errors
So as we really need this data we are thinking of options, how we could solve the problem by maybe adjusting Jira to store the data in a Datewarehouse-friendly way. First thought was for example to have a nightly or weekly script that scans over the epics and just writes the sum of the remaining efforts of all children and the spent efforts into a custom field - meaning we would have a history on an Epic level.
Do you think this is a suitable solution? Are there other options we currently do not see and which might solve our problems? How are others solving this problem, are there any best practices?
The use case of the tracking of the reorganisation of the epics seems to be a bit out of the limits of what currently is possible with the standard model of eazyBI. The scope changes can be better tracked with eazyBI if you could replace epics with some single choice custom field, allowing to import the changes in the dimension.
Otherwise, any custom data warehouse solution would be needed, specifically designed for this use case. Perhaps, eazyBI can be used for importing the data incrementally if you can provide the data of remaining effort through SQL query or REST API:
Hi @janis.plume, in the meantime we solved the problem by having a “Cron-Job” (Scriptrunner) on our Jira instance, that periodically runs over all Epics and calculates two Custom Fields (Aggregated Remaining and Time spent). This gives us a history that we can import into eazyBI, although it creates some redundancy of information in Jira. But this report has a useable load time (seconds compared to minutes) and a granularity with which we can work.
Kind regards
Tobias