I am new to this and having read the Newbie FAQ's etc I am trying to resist the temptation to fiddle with BOINC priorities. However, I have received virtually no work since 18 November (nothing from Rosetta in the queue at the moment) and also notice that even when there were several jobs in the queue, SETI had taken both my processors for itself, leaving Rosetta in the cold! Should I just continue to leave it alone to find its own balance or should I fiddle?

(Running dual-core intelMac, so expected SETI and ROSETTA to run in parallel...)

Gerry

---

I am new to this and having read the Newbie FAQ's etc I am trying to resist the temptation to fiddle with BOINC priorities. However, I have received virtually no work since 18 November (nothing from Rosetta in the queue at the moment) and also notice that even when there were several jobs in the queue, SETI had taken both my processors for itself, leaving Rosetta in the cold! Should I just continue to leave it alone to find its own balance or should I fiddle?

(Running dual-core intelMac, so expected SETI and ROSETTA to run in parallel...)

Gerry

Hi GerryMCSHERRY

Please tell us some more about how your settings are for the projects.

And if there are any messages in your BOINC about why you don´t get any work.

Anders n

---

Hi Anders N...Thanks for picking this up.

Can you see my computer and preferences...? If not which bits are most useful to paste into a reply?

No messages relating to supply of work. I have tried a few manual updates of Rosetta but not getting any extra WUs, meanwhile SETI trundles along and keeps resupplying itself!!

Gerry

---

what are your resource shares? I see you attached to Rosetta on the 15th of November. I'd wait and see what Boinc does to balance your requested resource shares. It's only been 4 run days and the scheduler doesn't keep the balance on periods that short, it maintains the balance long term. The work scheduler depends on so many factors that it'd be easier to ask you to read about the "work scheduler" in the Wiki.

Welcome to Rosetta

---

Hi Astro...

Resource share is 100 for each of SETI and ROSETTA. I assume that means equal shares across both processors. When I first started it was very regularly 1 processor busy on each application for the first few days...it's just in the last day or so that SETI seems to have taken over both processors and now I only have SETI WUs in the queue.

Perhaps I need to take your advice and chill for a few days until it settles down?

Gerry

---

I would. if you look up the "work scheduler" you'll see Boinc doesn't use "CPU affinity" in it current release. Boinc keeps track of "cpu seconds" and balances that. The larger you cache of total work, the funnier Boinc will "seem" to run. The shorter your cache the more likely you are to see it working like you "think" it might. This is because of EDF (earliest deadline first), where boinc thinks "hey, I might miss a deadline so I'll run wus based upon deadlines, then when it's not in fear of missing a deadline it'll switch back and forth. There are many other factors, which is why I linked you to the work scheduler in the wiki. I have a cache of 1/2 day. To keep it from "looking" funny, I'd recommend keeping your "connect to" setting less than 2 days.

If you've read and understand the work scheduler, there is a small program called BoincDV (debt viewer) which you can use to look at the debt. All debts can be found in C:programfilesboincclientstate.xml.