Any Idea when milkyway will be creating new work. All my pc's have been out for about 24hrs

---

m4rtyn
******************************* *******************************

immediately

---

Dave Przybylo
MilkyWay@home Developer
Department of Computer Science
Rensselaer Polytechnic Institute

immediately

Thanks Dave!

---

Thanks Dave, they are coming down well. But the work available is likely to be mopped up quickly as PCs pick up again

Thanks Dave, they are coming down well. But the work available is likely to be mopped up quickly as PCs pick up again

*crunch crunch munch munch*

Milkyways just taste too fine... ;-)))

---

Lovely greetings, Cori

I think the 20 WU limit needs to be reinstated. I've just accidently been downloaded 80 WU's due to putting the point the wrong side of the 5 on connection time. All due back on the 08/11/2008.

michael

I think the 20 WU limit needs to be reinstated. I've just accidently been downloaded 80 WU's due to putting the point the wrong side of the 5 on connection time. All due back on the 08/11/2008.

michael

But the 20 WU limit is still working. It's 20 WUs per core so for a quad 80 would be just fine. ;)

---

Lovely greetings, Cori

Been getting 0 still. Just not getting it at the right time.

---

Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.

I'm not getting any WUs either. My box is crunching MalariaControl and Enigma atm, so it's not idle, but I hope MW will manage to distribute some more WUs soon :-)

Feed Me :)

Cori.

Thank you for the imfo. What sort of over run is there on WU even 24/7 I do not think they will finish by deadline date.

Michael.

Cori.

Thank you for the imfo. What sort of over run is there on WU even 24/7 I do not think they will finish by deadline date.

Michael.

If you run the "stock app" then maybe you won't finish the WUs in time (there are two WU lengths - one took me about 4 hrs and the other about 7 hrs on my dual core lappy).
But with the new optimized app by Milksop you will run every Wu within minutes - they will be cunched away very quickly. ;-)

The optimized app is posted here - with some instructions where to put it. ;-)

---

Lovely greetings, Cori

With the enhanced application there are two related issues. One, the vast rate increase means that a major increase in work units is being requested over time by each workstation -- so the MilkyWay folks need to generate a LOT more work units. Two, the cache maximum of 20 work units per CPU - which used to represent a few days of work for a CPU now represents less than 2 hours cache.

Strikes me that there are a couple of solutions out there. The first one would be revised work units which run longer (say 2 hours instead of 5 minutes) -- that would reduce the ongoing heavy I/O load on the server (by a factor of 24 which only partially offsets the optimized application 50 to 1 increase in performance). Doing this would also mean that a 20 work unit per cpu cache would be close to a two day cache - a much more reasonable cache.

The alternative approach would be for the project to generate a LOT more work units of the existing length and bump up the cache from 20 per CPU to 200 or so providing almost a full day cache. Personally the first approach makes a lot more sense from user perspective - 5 minute work units are simply too much work load I/O wise and have their own inefficiencies for processing.

In the interim, a work around for you would be to do some 'babysitting' the scheduling process -- as work units are being continuously generated, after you get a '0 work unit' downlad, hit the update button again, and again, and again and again. Definitely a pain and this does stress the server. But you can get work units this way.

I'm not getting any WUs either. My box is crunching MalariaControl and Enigma atm, so it's not idle, but I hope MW will manage to distribute some more WUs soon :-)

---

immediately

I guess we can't get any quicker than that! ;)
Thanks!
Rod

Good evening Cori.

Thanks again managed to get all 80 back. After a struggle. Running vista didn't help any. Trying to find hidden files.

Doing a 7 hour 33 miniute WU in six miniutes. I can see now why so many Boinc projects run out of work so often.

Michael.

Well that's the deal -- it really wasn't a 7 hour work unit, it just took that long to run prior to the optimization. So we can hope for new style work units that run at least an hour -- make like simpler all around.

Doing a 7 hour 33 miniute WU in six miniutes. I can see now why so many Boinc projects run out of work so often.

Michael.

---

we need more work immediately.

Very true, but here other projects are getting crunched as well.

Indeed they are -- again, the effect of a 5 minute work unit means out of work scenarios become the norm. It is VERY apparent that a new class of work unit is required -- one which runs soemthing like 1 to 3 hours so that the load on the server is more reasonable and that caches can approach 1 to 2 days instead of less than two HOURS. Two hour caches (when work units are availalble) are WAY TOO SMALL for steady work. Said before, saying again.

Very true, but here other projects are getting crunched as well.

---

While I agree with most of what you have stated.

If I had stuck to stock appp the WU's on my Q6600 would have taken 7 hours 33 miniuts, and would have ment over running report dates.

Changing up a gear they then took 6 - 7 miniutes.

I to am now out of work.

But making the work units longer 4 hours would knowing human nature just get someone to design a new app that would reduce it down to 6 miniutes.

I have been running this type of project since seti started in 1999 and as WU get longer apps get faster,and host get OC.

IMHO The main motive behind this is credit hunger.

Michael