21) Message boards : Cafe LHC : Einstein is back up..... (Message 14558)
Posted 15 Aug 2006 by Gaspode the UnDressed
Post:

BUBO

I spy with my little eye something beginning with ... K

22) Questions and Answers : Sixtrack : KADIKÖY SU TESİSATI - İSKİ KADROSU (Message 14543)
Posted 14 Aug 2006 by Gaspode the UnDressed
Post:
Hi! This is an English language site - you might get a better response if you post in English.
23) Message boards : Team invites : No work here ? (Message 14424)
Posted 24 Jul 2006 by Gaspode the UnDressed
Post:
No work here - it's true.

LHC@Home is currently ina quiet phase. Studies aredone in batches, the resukts analysed, then more studies. At present the studies are quite small: a batch of work will run out in minutes, and take a week or two to crunch.

If the nature of the project interests you then stick with it and expect only sporadic work. If it's regular work you want, look elsewhere.

Oh - and there's no SysAdmin on the project right now, so problems take a long time to fix, if they get fixed at all.

24) Message boards : Number crunching : Duplicate post (Message 14420)
Posted 24 Jul 2006 by Gaspode the UnDressed
Post:
Duplicate post
25) Message boards : Number crunching : Bye all! (Message 14419)
Posted 24 Jul 2006 by Gaspode the UnDressed
Post:
I'm sorry to say, having been one of the earliest beta members that the current parlous state of this project leaves me totally disillusioned.

So - I'm closing down my BOINC activities.

It's been fun, sometimes.

Crunch well!

26) Message boards : Number crunching : I think we should restrict work units (Message 14380)
Posted 20 Jul 2006 by Gaspode the UnDressed
Post:

Remember when I said let the string go and the tennis ball will go straight? Well somewhere in here LHC lets the particle go and it smashes into the target. Same place every time. Think you can let go of the string and have the ball go the same direction every time? I hear you crunching again.


Not so. The beams do indeed run parallel within the same tunnel. There is no 'exit point'. Rather, at points around the ring the beams cross, and it is at these points that the real experimental science takes place.

As the beams cross there is an occasional collison. I say 'occasional' for a reason. Most particles in the beam just sail straight through and continue round the accelerator. 'Most' is very nearly all - maybe all of them most of the time. But, because of the immense speeds generated any given particle will cross the collison points millions of times per second, the possibility of a collision within a sensible time frame becomes workable.

There is a separate modelling program (Geant4) that models the behaviour of particles and their products at the collision point. This was the subject of a BOINC porting exercise, but the computing environment required will defeat most home computers.

27) Message boards : Number crunching : Why can't I get any work? (Message 14370)
Posted 19 Jul 2006 by Gaspode the UnDressed
Post:
Okey Doke...

Waxing analytically for a moment, I found the below on one of the About LHC pages...

So my question is, do I have a processor that LHC does not trust? I know I am slow, old, etc., and this may be a logical reason for not gettign work units. Or it could just be that I am VERY unlucky, or maybe it is a cache thing, so I am downsizing mine again just to see what happens. I will try .4, because I think that is what it was set at when I did get some stuff...

Any thoughts??

D


The libraries referred to in the extract you found have done a very good job of standardising results across processors. If your processor is untrusted then you'll get a message in your output file to the effect of 'There was work available for other platforms'. This is the sort of message you'd see if you had a Mac, for example.

If you're processor is just turning in bad results then you'd never get any credit. By degrees your 'Maximum daily WU quota per CPU' setting (listed against each computer) would creep down to zero, and you'd get 'Quota exceeded'messages.

If all you're seeing is 'No work from project' then you're being unlucky. Blame the crunchers with seven day caches for sucking the well dry before you get to it.
28) Message boards : Number crunching : Is the Validator Off-Line? (Message 14325)
Posted 14 Jul 2006 by Gaspode the UnDressed
Post:
All? Patience!

Of the three of your machines that received work on the 13th only one now has a result from a WU that is pending.

29) Questions and Answers : Preferences : Why can't I get any work? (Message 14300)
Posted 13 Jul 2006 by Gaspode the UnDressed
Post:
"If you don't want me in this project, just say so"

Why did you find it necessary to post your entire message file four times? If you don't want to play nicely please go and play elsewhere.

30) Questions and Answers : Preferences : Why can't I get any work? (Message 14299)
Posted 13 Jul 2006 by Gaspode the UnDressed
Post:

"If you don't want me in this project, just say so"

Why did you find it necessary to post your entire message file four times? If you don't want to play nicely please go and play elsewhere.

31) Message boards : Number crunching : I think we should restrict work units (Message 14286)
Posted 10 Jul 2006 by Gaspode the UnDressed
Post:
Bored, bored, bored.

Has anyone anything new to say?

32) Message boards : Number crunching : LHC@home Alpha? (Message 14240)
Posted 5 Jul 2006 by Gaspode the UnDressed
Post:
The alpha is as the names implies an alpha project. I used it to test updates and newer versions of sixtrack, before moving them to the main server.
...
I guess it is also the first place they will try out the garfield application once it is done. It has run on the cern internal boinc system and that worked. I think. I can tell you more once i see the meeting minutes.


Please, can you send me e-mail, when they accept more alpha testers? I just want to join that project. My e-mail address is henri.heinonen@mbnet.fi.


It isn't wise to post your email address on a public site. I hope you have a good spam filter.

33) Message boards : Cafe LHC : Holiday Weekend in the U.S.A. (Message 14237)
Posted 4 Jul 2006 by Gaspode the UnDressed
Post:

[cheeky wink] Possibly because our history goes too far back for us to need reminding that we have a history [/cheeky wink]





Happy St George's Day!


(23rd April, for our american cousins...)
34) Message boards : Cafe LHC : Holiday Weekend in the U.S.A. (Message 14234)
Posted 4 Jul 2006 by Gaspode the UnDressed
Post:
I didn't see the Cross of St George appearing here in April, John. Why was that?
35) Questions and Answers : Windows : any download (Message 14220)
Posted 1 Jul 2006 by Gaspode the UnDressed
Post:
There's no work available at LHC at present. LHC@Home issues work in a batches. Each batch studies particular characteristics of the machine. Once the results are in they are analysed, and the results of this analysis determine the parameters for the next study. It is normal for there to be a break of a week or two between studies while the analysis is done.


36) Questions and Answers : Windows : Computer too slow for project? (Message 14214)
Posted 30 Jun 2006 by Gaspode the UnDressed
Post:
I'd guess that you might complete a full result in around four days, so you should be OK. However, it is the nature of LHC that some results (where the beam becomes unstable) can complet very much quicker.

Try it and see - you may be pleasantly surprised!

37) Message boards : Number crunching : I think we should restrict work units (Message 14209)
Posted 28 Jun 2006 by Gaspode the UnDressed
Post:


We are all arguing round in circles now...



Just like the nuclei in LHC.

Yes - we have become CERN's biggest accelerator to date!

Keep it up!

38) Message boards : Number crunching : I think we should restrict work units (Message 14199)
Posted 27 Jun 2006 by Gaspode the UnDressed
Post:

If Project Admin thought it was broken, they would fix it.


Except, of course, there is no project admin.

39) Message boards : Number crunching : I think we should restrict work units (Message 14195)
Posted 27 Jun 2006 by Gaspode the UnDressed
Post:
It is clear that 1000 results spread between 1000 machines will complete faster than 1000 results spread between 20 machines. In this regard, large caches wil slow completion of a study.

It is also true that where some users are using large caches they may take a disproprtionate amount of the available work and credit in a batch project like LHC. In this regard they could be construed as 'greedy'. However, all participants can run other projects if they wish. Credit is awarded by all projects, so for those crunchers unable to run LHC presumably they can use that time to run something else and earn credit that way.

Given the pattern of this project (i.e. study, pause, analysis, next study, pause...) it it reasonable to suggest that shortening the study time by spreading the work more evenly will improve the overall provision of work. However, one key factor mitigates against this: there is only one physicist submitting work. It is unlikely that she/he has nothing else to do while a study is in progress, so reducing the time required for a study won't necessarily reduce the time between studies.

Of course, there is one joker in the pack! Those with large caches will generally return results more slowly than those with smaller ones. This means that their results will generally return later, and are more likely to encounter the infamous file_deleter bug where valid results are never granted credit.

So, for those who really want a large cache, be prepared to do more work for less reward!

40) Message boards : Number crunching : Strange CPU variations (Message 14190)
Posted 26 Jun 2006 by Gaspode the UnDressed
Post:
I'm not sure there is a problem. LHC writes checkpoint data every so often, and while it does so it's waiting on Windows to write the data to the disk. This might account for what you see.


Previous 20 · Next 20


©2024 CERN