Hi Steve and thank you a lot for your replies. I read (I think most of ) your papers and I find them very precise and useful. Unluckily here I'm the only reader of them.
I have 2 question I'd really like to ask to you as WLM expert. I received answers, but I ask them again: 1) Say I have a velocity goal of 90% and importance 1, PI is at max 75% and the workload NEVER achieves this goal.Never. What's WLM behaviour ? Does it continue to serve work to try to reach the goal of 90% in a endless manner (so consuming CPU )or, after a while, it stops to serve it ? And if yes, in which way ? Reducing the goal ? Putting the work in another service class ? Lowering Dispatching priority ? I made this question in a seminar and the answer was it stops to serve this service class, but I don't know what service class is assigned in this case or how WLM behaves. 2) If I submit a job and it is waiting to be submitted in JES queue (say there's no init available) does WLM consider it so it could be in a different period (say second period ) even if it is still waiting (or started after a while) ? I was told that's the case, ie WLM start to consider the job after it's submitted and not only if it is active and running. Thank you in advance and best regards Max Scarpa ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
