Best Practice: Recurring Tasks

SUGGESTED

I am seeking best practices as it pertains to recurring tasks. Has anyone built a way to schedule recurring tasks based on criticality and dependency? The common theme here seems to be related to it takes too much time or it never completes. In theory you would need to know and understand just how long a task takes to complete in order to properly set the times etc... We built a report in SEI which list all the parameters of the tasks etc... because we have over 185 tasks and yes this is far too many! I would like to know if there is a way to flag a task non-critical and if the system is constrained just don't run the non-critical task of course without someone having to go into SAGE and shut the task down. Tasks like allocations in my mind are not critical if a PC planner or shipping agent can just hit the allocation button etc... yet they are clogging the system. I guess what i am asking is, are there best practices and is there already a list if you will of task dependencies? 

Any help would be greatly appreciated. 

  • 0

    Hi Heather,

    To get the discussion started...The first thing might be to get with your users and find out what each task means to them. Why are they running it and why does it have to run that often or at that time? What is the advantage to them of running the task as a recurring task and not running it manually?

    Also, what version and service patch are you currently on?

  • 0 in reply to pnightingale

    Hi. We are on V12 patch 27. I have been asking the business for the dependencies in some cases they can provide. In others they come to "us" to solve. We only recently upgraded from V9 to V12 and very simply copied what was in V9. Sadly we are plagued with performance issues in V12 overall and recurring tasks was one of my staring points. My intent was to see if there were any best practices out there already as to avoid recreating the wheel if you will. lol Thank you for your reply!

  • 0

    Ok so we have taken a look at all of our recurring tasks and tried to resolve any potential conflicts. Of course the performance issues ebb and flow which I am sure others experience. That said we are in the midst of investigating all the recurring tasks and checking for timing etc... What I would like to ask is some of the tasks are grouped if you will. Which should really be ok. We are wondering if there is any gain in splitting the batch to maybe All US, MX, and Europe to run during Asia up time and create a secondary batch for our Asia region to execute during US, MX, Europe time. Would there be any benefit or performance gain if we swapped or split the batch and run it during the downtime for each respective region? Basically in simple terms when US, MX and Europe runs during Asia hours and  we run Asia during US, MX and Europe hours. We are still finding the request getting stopped or building so many records it will take a long time to flush out. In particular our accounting tasks are both the victim and the culprit. Any help would be appreciated. 

  • 0
    SUGGESTED

    To chip into discussion, I also have this issue.

    Try Application Clustering (https://support.na.sage.com/selfservice/viewContent.do?externalId=109651&sliceId=1)?

    Last I remember it will need a brand new VM/server instance for every new main runtime and has to have direct connection to main Application folders.

    I believe there's no harm to try but according to docs, just make sure the runtimes are all up at all times and leave it to Sage X3 batch server to decide which runtime to do it.

    In Batch Server, remember to adjust the max active query once the new additional main runtime is up.

  • 0 in reply to chunheng

    Thank you!! I will take a look. I have also found some posts on the UK boards you might want to take a look at  https://www.sagecity.com/gb/sage-x3/b/sage-x3-uk-support-insights/posts/batch-server-groups-of-tasks  Let me know if you can access. Worse case I can send the link another way perhaps.