Prev
Next
Forums Integration / Customization IPA/ProcessFlow

Process flow performance issues in Cloud - possible grid issues???

 4 Replies
 0 Subscribed to this topic
 52 Subscribed to this forum
Sort:
Author
Messages
Jessica M
Advanced Member
Posts: 34
Advanced Member
9/20/2017 3:17 PM

    We are getting ready to go live in the cloud on 9/22/17.  Our grid was tuned in the spring and our process flow performance  was better in the cloud than our on-prem version as of mid summer.  Now, all of the sudden, our performance has gotten worse.  Flows are running anywhere from 30 minutes more to 4+ hours longer than they previously were in the cloud. 

    We are not getting anywhere with our project team and the support personnel they have assigned.  At this point the answer has been "hmmm I don't know why" to "it must be a fluke".  I know on-prem they had to turn on S3 pooling.  However, in the cloud S3 is separate and that is running fine.  It's just our Landmark side that is performing poorly. 

    Does anyone have any suggestions?  Anything we should be looking at on the landmark grid?

    We would greatly appreciate any and all suggestions at this point. 

    Thanks,

    Jessica

    brox07
    Basic Member
    Posts: 8
    Basic Member
    9/20/2017 3:31 PM
      I'm not an expert in regards to this, especially cloud-based because our servers are internal, but could it be something as simple as an ISP speed issue? Have you tried running a speed test and seeing what Mbps you're getting?
      Jessica M
      Advanced Member
      Posts: 34
      Advanced Member
      9/20/2017 3:37 PM
        I am not sure that has been tested.  Our S3 side performs better than on-prem so I would think that if it were an ISP issue it would affect all areas.  Would you agree?  Or am I overlooking something.  I can certainly suggest this to our network engineers and our Infor support.
        brox07
        Basic Member
        Posts: 8
        Basic Member
        9/20/2017 3:46 PM
          Rereading your initial post, I'm wrong.  I thought you were saying the cloud-based servers were running slowly.  My bad.  We've had several instances of issues similar to this.  We had a couple of flows that were running super inefficiently, we also had to tune and change our LPA node configurations several times.  I think we're currently running three LPA nodes and trying to load balance across them.  I don't think Infor recommends that but it's seemed to work for us for a while now. 
          Jessica M
          Advanced Member
          Posts: 34
          Advanced Member
          9/20/2017 5:40 PM
            Thank you! We are not allowed to touch (or even see) our LPA nodes as we are hosted in the cloud. I am going to make that suggestion, but I am afraid since Infor has control of our grid/nodes they may not allow it.