Posts by Desti

21) Questions and Answers : Unix/Linux : FreeBSD 6.0 - unknown platform (Message 6892)
Posted 20 Dec 2005 by Desti
Post:
Rosetta does currently not support *BSD.
http://boinc.bakerlab.org/rosetta/apps.php

Maybe you could try the linux application with the anonymous plattform support.
22) Message boards : Number crunching : client for AMD64 X2 ?? (Message 6183)
Posted 13 Dec 2005 by Desti
Post:
Bill. What is the statistical likelyhood that a copy of the app_info.xml from a P4 machine running Linux would help?


I would guess that the app name is the same - worst case is try it and the results start erroring out instead of not being downloaded... I think you can just paste the file in here, and algon can take it from there. I get the impression he's comfortable at the bit-twiddling level.


OK, will take care of it tomorrow when I hit the office. Can't get at the box right now.


Ugh - now I'm confused. Nowhere in the BOINC directory hierarchy is there a file whose name matches 'app*'.

Help?


The app_info.xml is not automaticly generated or downloaded, it's a selfmade file for selfmade applications.
23) Message boards : Number crunching : client for AMD64 X2 ?? (Message 6159)
Posted 13 Dec 2005 by Desti
Post:
Hi,

you can

a) simply use a 32 Bit binary of BOINC
b) read this:
http://boinc-doc.net/boinc-wiki/index.php?title=Found_app_info.xml%3B_using_anonymous_platform
http://boinc-doc.net/boinc-wiki/index.php?title=BOINC_Anonymous_Platform_Mechanism
24) Questions and Answers : Unix/Linux : Won't upload WU (Message 4838)
Posted 30 Nov 2005 by Desti
Post:
There is no x86_64 application, you need to use a x86 BOINC client or read this http://boinc.bakerlab.org/rosetta/forum_thread.php?id=301#4781
25) Message boards : Rosetta@home Science : AMD64 Support? (Message 3929)
Posted 22 Nov 2005 by Desti
Post:
I'd think it would be nice to have 64bit versions. As the application is pretty new I would expect that this shouldn't be much work, but basically be a recompile only.


I can't speak for Rosetta, but I know several people have at various times created 64 bit versions of the SETI app, and the basic opinion has always come back to "there is no point". There is no speed gain, because floating point is floating point and isn't helped. It requires yet another version of everything be maintained. 64-bit processors can run 32-bit applications perfectly well. If an application gains something by being compiled for a 64-bit path, great, but if the only reason to do it is for the 0.000001% of the users who have turned off 32-bit support...


It's not that FLOP calculations don't benefit, it's that the SETI program itself doesn't use the type of code to benefit from a 64 bit app.

Rosetta may or may not, it depends on the algorythms.



Most applications don't benefit from the 64 bit adress space, but from the ability to use more registers in the 64bit modus (for example media encoding) http://www.linuxhardware.org/article.pl?sid=05/02/24/1747228&mode=thread
26) Questions and Answers : Unix/Linux : Some errors with 4.77 (Message 3927)
Posted 22 Nov 2005 by Desti
Post:
Hmm... All of a sudden it works...
Already two WUs without errors...



It's cold today, maybe you should check your cooling ;-)
27) Message boards : Number crunching : Is there a Solaris application in the works? (Message 3762)
Posted 20 Nov 2005 by Desti
Post:
there will be no sooner a solaris version (due to lack of equipment when i remember correctly) than the source code is made available, which was delayed due to cheating issues. you might watch the "opening the source code" (out of my head, might be named slightly different) thread to get the latest news.


http://boinc.bakerlab.org/rosetta/forum_thread.php?id=349 ;-)
28) Questions and Answers : Unix/Linux : cannot attach client to rosetta project (Message 3761)
Posted 20 Nov 2005 by Desti
Post:
hmmm...courious, is this your system you wanna attach? http://boinc.bakerlab.org/rosetta/hosts_user.php?userid=10139

Try to attach with boinc:

boinc -attach_project http://boinc.bakerlab.org/rosetta YOUR_ACCOUNT_KEY

29) Questions and Answers : Unix/Linux : Some errors with 4.77 (Message 3635)
Posted 18 Nov 2005 by Desti
Post:
Do you use the self made Gentoo BOINC or the official binary?


I tried the official BOINC-Bin and my own build BOINC (no ebuild - from CVS - stable revision)

But BOINC should not have to do with that, right?
I mean: BOINC is just distributing and monitoring the WUs...
The Rosetta-Application itself is processing the WUs, right?


Right.
But this "execv: Text file busy" doesn't look like a typical Rosetta error.

For the glibc errors, what CFLAGS do you use for your system?
30) Questions and Answers : Unix/Linux : Some errors with 4.77 (Message 3621)
Posted 18 Nov 2005 by Desti
Post:
Do you use the self made Gentoo BOINC or the official binary?
31) Message boards : Number crunching : Processor Speeds (Message 3343)
Posted 15 Nov 2005 by Desti
Post:
Isn't the celeron M based on the pentium m core? If so, it's closer to the amd archetecture (shorter pipeline) than the p4. The pentium m does more work per mhz than the p4, just like the modern amd chips.


That's right. The Celeron M is a Pentium M with half L2 cache size and some less power saving steppings.
32) Message boards : Number crunching : code release and redundancy (Message 2789)
Posted 10 Nov 2005 by Desti
Post:

I've argued against redundancy in the past because it is a waste of resources. But perhaps we should go to two fold redundancy because of the credit issue. How about this: when we get to 1,000,000 credits per day, we go to two fold redundancy and give out the code. (Anybody want to place a bet on when we break the 1,000,000 a day mark?)

what do people think?


I for one would probably leave the project if I knew that the exact same work my computer was doing is duplicated be someone else. It feels like being distrusted and that my contribution is not really needed. Your suggestion to go to twofold redundancy when 1,000,000 credits per day are reached also suggests to me that the project doesn't really need that much computing power, another reason to leave the project at that stage.

Also, regarding the cheating issue, since currently the gained credit is donated CPU time x benchmark rating, it isn't possible to get more credits by tinkering with the rosetta code, or am I missing something here ? So why should giving out the rosetta code be an issue as far as cheating for credits is concerned ? Since the energy can easily be re-calculated from the returned structure I also don't see why redundancy should be required to detect invalid results.

-H.B.



Even with the progress of the last two days, it is clear we are going to be CPU time limited for the forseeable future. So redundancy really is wasteful. How would people feel about this: we do not resort to redundancy, so every calculation is unique, but we wait to release the code until the credit issue is resolved.



Sounds good.
What license will Rosetta use?
33) Questions and Answers : Unix/Linux : Unstoppable Rosetta client... (Message 2764)
Posted 9 Nov 2005 by Desti
Post:
Sounds like Rosetta wents into a zombie process.
34) Message boards : Number crunching : Linux vs Windows point awards (Message 2718)
Posted 9 Nov 2005 by Desti
Post:
An interesting thread on the seti boards: http://setiathome.berkeley.edu/forum_thread.php?id=22079


So if Rosetta started to count flops there would be less of a variance :)


EDIT: Also found this post on seti, from David Baker


That'a good idea. I hope they will introduce it soon. :-)
35) Message boards : Number crunching : Congrats to the rest of Free-DC for hitting #1 team spot (Message 2603)
Posted 7 Nov 2005 by Desti
Post:
Wooooohaaaa!


We are...erm...nearly close behind you.
36) Questions and Answers : Unix/Linux : WU's fail with several error codes (Message 2506)
Posted 6 Nov 2005 by Desti
Post:
Take a look into the BOINC Wiki http://boinc-doc.net/boinc-wiki/index.php?title=Main_Page
37) Message boards : Number crunching : Find a Drug is heading this way! (Message 2390)
Posted 6 Nov 2005 by Desti
Post:
Welcome to Rosetta and have fun!
38) Message boards : Number crunching : Linux vs Windows point awards (Message 2012)
Posted 1 Nov 2005 by Desti
Post:
Yea, this subject has pretty much been beaten to death.


That's true, but normaly the validation process gives the Linuxers mostly more credits than they claimed for. Rosetta has no validation against Windows(tm) crunchers, so there is no credit mixing.
39) Questions and Answers : Unix/Linux : Unstoppable Rosetta client... (Message 1912)
Posted 29 Oct 2005 by Desti
Post:
Do you use BOINC 4 or 5?
40) Message boards : Number crunching : New type of WU? (Message 1644)
Posted 23 Oct 2005 by Desti
Post:
Three point WU anyone?


Luxury!

2.35


2.20


Previous 20 · Next 20



©2024 University of Washington
https://www.bakerlab.org