Message boards : Number crunching : LHC@Home is NOT a science project
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Gaspode the UnDressed

Send message
Joined: 1 Sep 04
Posts: 506
Credit: 118,619
RAC: 0
Message 9772 - Posted: 31 Aug 2005, 16:50:54 UTC
Last modified: 31 Aug 2005, 16:51:12 UTC

Let's get one thing straight in all this whining about deadlines: LHC@Home is NOT</B> a science project. The science starts in 2007. Meanwhile LHC@Home is supporting an engineering project the like of which has never been seen.

In 2007 CERN will switch on the most powerful machine mankind has ever built. The computing we are doing now has to be done on time, and it has to be right. If it's not, at best, a large amount of money will be wasted waiting for us. At worst, the most powerful machine ever built has the potential to explode with the loudest bang ever heard.

If the project team has to adjust the parameters to meet project deadlines then so be it. It's not about credit. It's not about little green men. It's about getting things done in the [b]real world
.

I agree - the deadlines now are tight. I have adjusted my settings to meet the requirements. If other users don't wish to run to these constraints then this isn't the project for them, and they should find another. That's OK, and without sarcasm I wish them well.




Gaspode the UnDressed
http://www.littlevale.co.uk
ID: 9772 · Report as offensive     Reply Quote
J D K

Send message
Joined: 27 Aug 05
Posts: 50
Credit: 24,055
RAC: 0
Message 9794 - Posted: 1 Sep 2005, 4:20:13 UTC

Right on Bro, tell it like it is.........
ID: 9794 · Report as offensive     Reply Quote
Profile FZB

Send message
Joined: 17 Sep 04
Posts: 23
Credit: 6,871,909
RAC: 12
Message 9796 - Posted: 1 Sep 2005, 5:00:24 UTC

agreed, though i argue on "the loudest bang ever heard" part... ;)
ID: 9796 · Report as offensive     Reply Quote
Travis

Send message
Joined: 13 Jul 05
Posts: 7
Credit: 182,621
RAC: 0
Message 9797 - Posted: 1 Sep 2005, 5:11:26 UTC

Agreed
ID: 9797 · Report as offensive     Reply Quote
Profile The Gas Giant

Send message
Joined: 2 Sep 04
Posts: 309
Credit: 715,258
RAC: 0
Message 9799 - Posted: 1 Sep 2005, 5:31:16 UTC

Oh please give me the wu scheduling of BOINC 4.19 (pure resource share) and let me worry about whether a wu will be completed before the deadline (I never missed one yet where I had the ability to upload it), but the project and wu management (suspend/no new work) of BOINC 4.72 and the wu estimation calculation of 4.72 (it's better than nothing - since the raw numbers are sooo bad) to ensure I get sufficient wu's to last at least 72hrs (did I say some of my machines are on dial-up equivalent?). Give me a BOINC where the EDF mode starts at more than half the earliest deadline, let's say 80% of the earliest deadline.

That's all I care about.

Remember Pirates (aaarrrgggghhhh) ? They had deadlines that were very short indeed, in the realms of hrs not days.

Is it science or engineering...who cares? All I know is we have an overall project deadline and BOINC has returned more work using sixtrack than any other means available at CERN and will continue to do so AND this was without being on much at all. This has enabled CERN to do studies that they never thought they would have the time for....perfect don't you think!

Damn...appointment has arrived gotta go. Might have to finish this later.

Live long and crunch.

Paul
(S@H1 8888)
BOINC/SAH BETA
ID: 9799 · Report as offensive     Reply Quote
Profile JigPu

Send message
Joined: 1 Sep 04
Posts: 26
Credit: 600,998
RAC: 0
Message 9802 - Posted: 1 Sep 2005, 6:27:56 UTC
Last modified: 1 Sep 2005, 6:28:25 UTC

Amen.

If you're computer can't cope with the deadlines, by all means suspend or detach from the project (it's a lot nicer than being forced to micromanage!). However, assuming you're not on dial-up, I'd also try figuring out why BOINC isn't doing it's job correctly (and earliest-deadline-first mode IS correct functioning if it returns results before deadline ;)) since it should be quite capable of completing work before it's due (especially with LHC's extrodinarily incorrect predictions, BOINC's built in 80% fudge-factor, and the notoriously low system-uptime/boinc-uptime numbers BOINC generates).

Puffy
ID: 9802 · Report as offensive     Reply Quote
AnRM

Send message
Joined: 14 Jul 05
Posts: 13
Credit: 424,554
RAC: 0
Message 9803 - Posted: 1 Sep 2005, 6:50:44 UTC

Amen to our Gassy friend.....I'm still using 4.19 and will keep doing so as long as possible. Boinc 4.19 is a perfect example of the KISS principle: it is simple and has low CPU overhead and you know what to expect. I guess I would be a Luddite in an earlier epoch.....Cheers, Rog.
ID: 9803 · Report as offensive     Reply Quote
Profile adrianxw

Send message
Joined: 29 Sep 04
Posts: 187
Credit: 705,487
RAC: 0
Message 9804 - Posted: 1 Sep 2005, 8:14:33 UTC

I use 4.25. It has a few advatages over 4.19 but none of the newer pollution. I know how to manage my computer, BOINC never will.

As to the thread in general, presses + button.

Wave upon wave of demented avengers march cheerfully out of obscurity into the dream.
ID: 9804 · Report as offensive     Reply Quote
Profile ksba

Send message
Joined: 27 Sep 04
Posts: 40
Credit: 1,742,415
RAC: 0
Message 9827 - Posted: 1 Sep 2005, 15:58:25 UTC - in response to Message 9772.  
Last modified: 1 Sep 2005, 16:00:41 UTC

<blockquote>things done in the real world.
</blockquote>

Grumpf :-)
Agreed, but i'm still unhappy :-)
Now some of my maschines, that not run the whole time need to switch to more "unsexiest" projects. But you're right and i'm happy to help... with less power. ;-)

[btw, i use 4.25 too, its the 'best-of-both-world']

btw is there a project with week's as deathline, that i can run on maschines that only run hours per week ... ?
ID: 9827 · Report as offensive     Reply Quote
Profile FZB

Send message
Joined: 17 Sep 04
Posts: 23
Credit: 6,871,909
RAC: 12
Message 9829 - Posted: 1 Sep 2005, 16:36:45 UTC - in response to Message 9827.  

<blockquote>btw is there a project with week's as deathline, that i can run on maschines that only run hours per week ... ?</blockquote>

other than cpdn i can think of none, and i have doubts that you finish a wu in the deadline range if the pc is only on hours in a week.
ID: 9829 · Report as offensive     Reply Quote
Travis DJ

Send message
Joined: 29 Sep 04
Posts: 196
Credit: 207,040
RAC: 0
Message 9836 - Posted: 1 Sep 2005, 22:40:27 UTC
Last modified: 1 Sep 2005, 22:46:30 UTC

I say this knowing how busy the programmers are at LHC@Home - but appearances do say a lot for how well their user base accepts changes. There often is little news on the front page that has clear and easy-to-understand meaning. If news had trickled out on the home page like this...

2005-08-25: We are thinking about decreasing the deadlines to 5 days in order to decrease turnaround time and increase our productivity. This way we can offer more workunits because we can analyze more complete production runs since they will complete sooner than in the past. Let us know what you think in the message boards

2005-08-30: The deadline has now been decreased to 5 days. Thank you for supporting our ideas, we value your input.


... this thread would have probably not appeared and the user base would have known what's going on without having to figure it out the hard way as it seems we have had to become accostomed to lately. I've often found in my experience that programmers are more the type to get things done and interact with the fewest number of people in the process - PR is not in most programmer's genetic code. :)

To sum this up, it would be nice if 'thoughts and considerations' would make it to the news page in addition to real project news .. it allows the user base to know what the programmers are thinking and have in store, be prepared for changes, and stay informed. I believe a great number of you would agree with me the most "news" appears in the LHC message board rather than the front page where it belongs.

ID: 9836 · Report as offensive     Reply Quote
Profile Contact
Avatar

Send message
Joined: 17 Sep 04
Posts: 52
Credit: 1,699,607
RAC: 37
Message 9837 - Posted: 2 Sep 2005, 1:04:32 UTC

Hi MikeW. I must confess to being a whiner. While I was doing the happy dance from Einstein@Home’s decision to allow my slow dial-up relics to complete their work with 14 day slack, I saw LHC@home deadline dramatically reduced.

Travis DJ you very correct. Warning would have helped. Your frontpage news should be used as a template!

But the deeper story here is that I see many BOINC projects with very short deadlines.
I can see BOINC in the future being reduced to crunching 1 project at a time.
This greatly reduces her beauty.

I don’t see the devastation of allowing 7 days of slack. All projects should be encouraged to allow deadlines of 14 days or more. The hoarders and the slow-pokes will slow things down a bit, but in the long run of most projects it will be ok.
I do not have the will to boycott short deadline projects but I will intervene with manager to make sure longer deadline projects don’t suffer in the short term. That’s my little protest.
I admire LHC@home and many BOINC projects. I admire BOINC. We have a real chance of large progress!
ID: 9837 · Report as offensive     Reply Quote
Profile Andy Gray
Avatar

Send message
Joined: 28 Sep 04
Posts: 9
Credit: 804,488
RAC: 0
Message 9851 - Posted: 2 Sep 2005, 15:08:54 UTC

I agree with TravisDJ. A little more news would be welcome. We only seem to be told about server problems.
ID: 9851 · Report as offensive     Reply Quote
Profile rbpeake

Send message
Joined: 17 Sep 04
Posts: 99
Credit: 30,618,118
RAC: 3,938
Message 9855 - Posted: 2 Sep 2005, 15:45:30 UTC - in response to Message 9851.  

<blockquote>I agree with TravisDJ. A little more news would be welcome. We only seem to be told about server problems.</blockquote>

I also agree. Communication from the Project Team is critical in my opinion for all BOINC projects, since that is all that us crunchers are getting from participating in each project (other than credits, which are really "empty calories" compared with the nutrition of the science/progress/updates/news from the Project Team!).

Thank you, this would be much appreciated. :)
Regards,
Bob P.
ID: 9855 · Report as offensive     Reply Quote
paperdragon
Avatar

Send message
Joined: 2 Aug 05
Posts: 13
Credit: 190,657
RAC: 0
Message 9870 - Posted: 2 Sep 2005, 22:03:24 UTC - in response to Message 9827.  

<blockquote><blockquote>things done in the real world.
</blockquote>

btw is there a project with week's as deathline, that i can run on maschines that only run hours per week ... ?</blockquote>

On a Pentium 3 running at 933 megahetz I can do a predictor unit in 1 hour 15 minutes. So that might be your best bet, if BOINC is running 100% of the time that the machines are on. Predictor web site: http://predictor1.scripps.edu/

But since Predictor only has a seven day deadline, if you are only turning the machines on one day a week, then even this one would not work.


You like Myst? Uru Live returns! www.urulive.com
ID: 9870 · Report as offensive     Reply Quote
alpina

Send message
Joined: 3 Aug 05
Posts: 49
Credit: 143,072
RAC: 0
Message 9871 - Posted: 2 Sep 2005, 22:15:34 UTC

I also agree with many of the other crunchers here: communication is one of the most important stimuli to paticipate in a project. It's offcourse important that the project's subject is interesting and that it runs flawless but if you do not get any feedback(are little feedback) it all becomes a bit boring in the end.


BOINC.BE: The team for Belgians and their friends who love the smell of glowing red cpu's in the morning
ID: 9871 · Report as offensive     Reply Quote
Heffed

Send message
Joined: 2 Sep 04
Posts: 71
Credit: 8,657
RAC: 0
Message 9876 - Posted: 3 Sep 2005, 8:07:38 UTC - in response to Message 9772.  

<blockquote>Let's get one thing straight in all this whining about deadlines: LHC@Home is [b]NOT</B> a science project. The science starts in 2007. Meanwhile LHC@Home is supporting an engineering project the like of which has never been seen.</blockquote>
Engineering is a science...
ID: 9876 · Report as offensive     Reply Quote
Profile [B^S] thierry@home

Send message
Joined: 13 Jul 05
Posts: 26
Credit: 27,752
RAC: 0
Message 9878 - Posted: 3 Sep 2005, 8:48:27 UTC
Last modified: 3 Sep 2005, 8:52:04 UTC

Engineering is a science .... to turn, in general, science into money..... :-)

ID: 9878 · Report as offensive     Reply Quote
Profile Cori
Avatar

Send message
Joined: 13 Jul 05
Posts: 19
Credit: 35,859
RAC: 0
Message 9905 - Posted: 4 Sep 2005, 2:33:01 UTC - in response to Message 9829.  
Last modified: 4 Sep 2005, 2:36:12 UTC

<blockquote>btw is there a project with week's as deathline, that i can run on maschines that only run hours per week ... ?</blockquote>

You may try PrimeGrid formerly known as Message@Home (PrimeGrid Homepage), it now has a deadline of 14 days and WUs are finished rather quick. But be aware that there is only pre-registration of accounts at the moment. (You can create an account as usual, but you'll have to wait "in the queue" to receive the email confirmation with the account key. And due to latest outage and server/homepage move, this queue may be a little longer...)

Happy crunchin' ;-)
------------------------
Greetings from Cori
ID: 9905 · Report as offensive     Reply Quote
Profile Rytis

Send message
Joined: 2 Sep 04
Posts: 17
Credit: 22,605
RAC: 0
Message 9929 - Posted: 5 Sep 2005, 14:54:58 UTC

PrimeGrid: 318 activated, 665 total (347 waiting). Activation speed: 15/day. So i guess we have 3 week long queue :) Though it may clear out faster, I may activate more accounts per day.
PrimeGrid
Administrator
ID: 9929 · Report as offensive     Reply Quote
1 · 2 · Next

Message boards : Number crunching : LHC@Home is NOT a science project


©2024 CERN