4 Replies Latest reply on Jul 10, 2014 10:36 PM by markrbowles

    Asset-based Analytics

    markrbowles

      I'm using Asset-based Analytics and find it interesting that it is not capable of fail over. This creates a single point of failure scenario. Any others out there share this concern? Suggestions for fail over / redundancy?

        • Re: Asset-based Analytics
          Marcos Vainer Loeff

          Hi Mark,

           

          Don't forget that this is the first release of the product so there are still a lot of new features to be added on future releases. The product manager will have to prioritize the ones he thinks is important according to the feedback from the customers.

           

          Could you please describe with more details which feature do you want to add to PI Asset-based Analytics? The fact that you can backfill data in case there is an issue is not enough?

            • Re: Asset-based Analytics
              markrbowles

              Good point Marcos.  I'm very pleased that we have this tool now it is extremely useful.   I had not considered the back fill capability.  Thank you for pointing that out.   That should be enough to ensure analysis data integrity during scheduled patching scenarios when the service may be unavailable.  I suspect future releases will have redundancy like notifications.

                • Re: Asset-based Analytics
                  skwan

                  Mark:

                   

                  Failover did not make v1 as the feedback we got from the CTP and Beta cycles and vCampus Live! VoxPop was that customers would rather have an earlier release without failover since they have immediate use for the software.  This was a tradeoff that we had to make.  As Marcos already indicated, you can backfill any missed calcs and more importantly, the current release will do automatic backfill for up to 3 days when the PI Analysis Service is restarted.  This would happen during an Microsoft OS patch, for example.

                   

                  In addition, we also want to take this opportunity to re-examine high-availability features of other components of the PI Server, for example AF.  There are new technologies available now that were not available years ago so we're always looking for a better way to do things.

                   

                  In short, we're aware that failover (some form of high availability) is necessary.