• 1
    • Our product teams collect and evaluate feedback from a number of different sources. To learn more about how we use customer feedback in the planning process, check out our new feature policy.

      Atlassian teams continually monitor performance, looking at the time it takes to load the board for the first time and subsequent times.

      • First time: We call this "Initial Load", and it includes resources for all the views plus assets like images, so it is usually longer.
      • Subsequent times: We call this "Transitions", such as loading a board a 2nd time through clicking on the project side navigation after transitioning from the backlog (e.g. board -> backlog -> board), which is faster because most assets are already loaded.

       

      To improve performance for boards with 300-500 cards, we are aiming for several milestones as part of a longer journey:

      Today:

      • Initial Loads: 90% of board loads within 12s, and 99% within 34s
      • Transitions: 90% of board loads within 6s, and 99% within 13s

      Milestone 1:  Achieved as of July 2021

      • Initial Loads: 90% of board loads within 10s  (currently 7.6s), and 99% within 27s  (currently 25s)
      • Transitions: 90% of board loads within 4s  (currently 4s), and 99% within 11s  (currently 10.7s)

      Milestone 2:

      • Initial Loads: 90% of board loads within 7s, and 99% within 14s
      • Transitions: 90% of board loads within 2s, and 99% within 7s

      Milestone 3:

      • Initial Loads: 90% of board loads within 4s, and 99% within 10s
      • Transitions: 90% of board loads within 1s, and 99% within 3s

      We’ll update this issue with progress towards these milestones, and will do it at least every quarter. We have a public commitment to improve the performance of all key experiences within Jira (e.g. issues, search, filters, boards, and backlogs) as shown in our public roadmap.

       

      What about boards with more than 500 cards?

      We do not recommend having >500 cards on your boards. The board should ideally represent the highest priority for your team in the coming few weeks, and not contain all the issues in the project. This community post explains how to optimize your board to show < 500 cards.

            [JRACLOUD-85663] Make Next-gen boards with hundreds of cards snappier

            Hi all, we've achieved Milestone 1 as of July 2021. We'll be providing further updates in future as we continue focusing on improving the performance of bigger boards to reach Milestone 2's timings.

            Ivan Teong (Inactive) added a comment - Hi all, we've achieved Milestone 1 as of July 2021 . We'll be providing further updates in future as we continue focusing on improving the performance of bigger boards to reach Milestone 2's timings.

            61521ce46dba Thanks for your feedback! The decision to start another Jira issue is because the previous ticket was listed as a "bug" when it is actually not a bug. We have SLOs around bugs being fixed within a certain number of days, which will not be possible to meet as we need time to develop, test our experiments to prove and validate that they're a success. We also have to make sure that architectural constraints are resolved through overhauling our technical architecture. There are also user experience considerations that we need to review when optimizing for faster page loads (e.g. deferring something to make something faster will result in something else being slower). We still acknowledge that this is a problem for our customers, and we're definitely focusing on it.

            We can go on forever about what a bug definition is (in my opinion, it is something that was built in a certain way but it wasn't executing the way it's built), but I don't think that this is the case here. I think we both agree that we don't want to rush into closing a ticket for the purpose of fulfilling SLOs. Having a feature suggestion ticket is more appropriate in this case, as it allows us to continually update customers on our progress without being rushed into closing the "bug" ticket. Hope that makes sense!

            Ivan Teong (Inactive) added a comment - 61521ce46dba  Thanks for your feedback! The decision to start another Jira issue is because the previous ticket was listed as a "bug" when it is actually not a bug. We have SLOs around bugs being fixed within a certain number of days, which will not be possible to meet as we need time to develop, test our experiments to prove and validate that they're a success. We also have to make sure that architectural constraints are resolved through overhauling our technical architecture. There are also user experience considerations that we need to review when optimizing for faster page loads (e.g. deferring something to make something faster will result in something else being slower). We still acknowledge that this is a problem for our customers, and we're definitely focusing on it. We can go on forever about what a bug definition is (in my opinion, it is something that was built in a certain way but it wasn't executing the way it's built), but I don't think that this is the case here. I think we both agree that we don't want to rush into closing a ticket for the purpose of fulfilling SLOs. Having a feature suggestion ticket is more appropriate in this case, as it allows us to continually update customers on our progress without being rushed into closing the "bug" ticket. Hope that makes sense!

            Jez added a comment -

            Hey Ivan, we're continuing to watch these threads with hope of further improvements.

            Transparently, I am somewhat puzzled about the decision to start a new issue... I don't think the original next gen board performance issues raised have been fully resolved and whilst closing off the other issue and starting this may work for Atlassian's internal purposes, it does nothing for our everyday experience as users of the product unless it transpires into actual improvement.

            I guess the tension here perhaps is... what is a problem versus what is something that just needs improvement.

            From a customer perspective, the current performance is still a problem for us, our legitimate use of the boards doesn't allow our users to work efficiently and with good flow. The very creation of this new issue is a clear indication that Atlassian knows (and agrees with) this, so issue juggling aside, hopefully there continues to be a strong focus on resolving the issue as a bug, not as a feature request.

            Hope all is well with you and your team! Please do let us know if there is anything we can provide further customer feedback on here.

            Jez

             

            Jez added a comment - Hey Ivan, we're continuing to watch these threads with hope of further improvements. Transparently, I am somewhat puzzled about the decision to start a new issue... I don't think the original next gen board performance issues raised have been fully resolved and whilst closing off the other issue and starting this may work for Atlassian's internal purposes, it does nothing for our everyday experience as users of the product unless it transpires into actual improvement. I guess the tension here perhaps is... what is a problem versus what is something that just needs improvement. From a customer perspective, the current performance is still a problem for us, our legitimate use of the boards doesn't allow our users to work efficiently and with good flow. The very creation of this new issue is a clear indication that Atlassian knows (and agrees with) this, so issue juggling aside, hopefully there continues to be a strong focus on resolving the issue as a bug, not as a feature request. Hope all is well with you and your team! Please do let us know if there is anything we can provide further customer feedback on here. Jez  

              e4bb9a47d132 Mark Z (Inactive)
              mhowells Mike Howells
              Votes:
              4 Vote for this issue
              Watchers:
              11 Start watching this issue

                Created:
                Updated:
                Resolved: