Is it the AB_CASP6 job that's so boinc'in huge or is it the t216 protein?

---

dag
--Finding aliens is cool, but understanding the structure of proteins is useful.

Is it the AB_CASP6 job that's so boinc'in huge or is it the t216 protein?

ALL xxxCASPxxx files I've seen are HUGE. Longest I've seen.

---

Add this signature to your EMail:
Running Microsoft's "System Idle Process" will never help cure cancer, AIDS nor Alzheimer's. But running Rosetta@home just might!
https://boinc.bakerlab.org/rosetta/

All AB_CASP6_* jobs I've seen sofar (not many) have been on the low side of demands, i.e. just 60-70MBytes memory working set (vs up to 160-170MB for the *largescale_large_fullatom* jobs)

Also a P4 completes a model in 10-15min (vs 1+hr of the bigger WUs).

So overall, I'd say they're on the small side (by Rosetta standards) sofar.

Maybe other CASP jobs are much larger, looking forward to take advantage of the extra 512MB I added to my PCs recently :-)

---

Best UFO Resources
Wikipedia R@h
How-To: Join Distributed Computing projects that benefit humanity

Hi: yes the T216 protein is pretty huge, as is obvious from the graphics and from the lower speed at which these jobs crunch (there's jsut more computation per step). The download files for these are also big. But I think Dimitris is right, the memory footprint should be significantly less than the infamous largescale_large_fullatom jobs. In any case I've cancelled the workunits and sent them out again with a minimum memory requirement of 300 Mb.

One other point worth re-iterating: these jobs may take longer, but clients will get proportionally more credit!

All AB_CASP6_* jobs I've seen sofar (not many) have been on the low side of demands, i.e. just 60-70MBytes memory working set (vs up to 160-170MB for the *largescale_large_fullatom* jobs)

Also a P4 completes a model in 10-15min (vs 1+hr of the bigger WUs).

So overall, I'd say they're on the small side (by Rosetta standards) sofar.

Maybe other CASP jobs are much larger, looking forward to take advantage of the extra 512MB I added to my PCs recently :-)

---

Hi: yes the T216 protein is pretty huge, as is obvious from the graphics and from the lower speed at which these jobs crunch (there's jsut more computation per step). ... In any case I've cancelled the workunits and sent them out again with a minimum memory requirement of 300 Mb.

300 MB might be too low. On my box it takes up to 220 MB physical and 460MB virtual and one model needs more than an hour to finish (1GB RAM, AMD 64 2400MHz).

I assume one can't set a minimum processor requirement but I would suggest 513 MB minimum memory requirement. Thus it goes only to more recent hosts with >512 MB RAM (which tend to have a faster proc as well).

---