Message boards : Number crunching : donated time wasted
Message board moderation

To post messages, you must log in.

AuthorMessage
bass4lhc

Send message
Joined: 28 Sep 04
Posts: 43
Credit: 249,962
RAC: 0
Message 6927 - Posted: 10 Apr 2005, 2:08:38 UTC

many donators have wu's restarting at 0%
wasted cpu time. over and over again.

on the lhc@home site is no mention of this problem.

is our donated time that unimportant?

it is time you see us as donators. we pay to donate to your project.

information is all we ask.


ID: 6927 · Report as offensive     Reply Quote
Profile The Gas Giant

Send message
Joined: 2 Sep 04
Posts: 309
Credit: 715,258
RAC: 0
Message 6928 - Posted: 10 Apr 2005, 2:46:15 UTC

You must be blind...keep looking it's all there.

Paul
(S@H1 8888)
BOINC/SAH BETA
ID: 6928 · Report as offensive     Reply Quote
bass4lhc

Send message
Joined: 28 Sep 04
Posts: 43
Credit: 249,962
RAC: 0
Message 6930 - Posted: 10 Apr 2005, 2:57:48 UTC - in response to Message 6928.  

> You must be blind...keep looking it's all there.
>
thank you paul, in your opinion this might be a constructive answer.

if i was blind there is no way i could read this forum.
ID: 6930 · Report as offensive     Reply Quote
Profile Logan5@SETI.USA
Avatar

Send message
Joined: 30 Sep 04
Posts: 112
Credit: 104,059
RAC: 0
Message 6932 - Posted: 10 Apr 2005, 5:05:26 UTC
Last modified: 10 Apr 2005, 5:07:10 UTC

The sixtrack 4.64 DOES have the "zero credit" problem that has supposedly been corrected in the new 4.66.

I'll wager that all your "continued" problems are really nothing more then you're still crunching under the older version.

If so, you have 3 choices:

1) Be patient and wait. Your clients will eventually down the 4.66 sixtrack
2) Reset the project which will delete everything in your LHC folder and download new work under the 4.66
3) Move to another project if the admins and the project are not meeting your expectations. There's nothing keeping you here if you are as unhappy as you appear to be....

If you've bothered to look through the older posts, you'll see that the admins do NOT consider the "zero credit" WU's a waste as it's still the science being processed regardless, and that SHOULD be the most important reason to crunch, NOT the credits granted or the glory or whatever else drives a person.

Does this help clear things up for you?



ID: 6932 · Report as offensive     Reply Quote
Profile The Gas Giant

Send message
Joined: 2 Sep 04
Posts: 309
Credit: 715,258
RAC: 0
Message 6938 - Posted: 10 Apr 2005, 11:15:48 UTC

OK, I'll bight....have a read of these.

http://lhcathome.cern.ch/forum_thread.php?id=1229
http://lhcathome.cern.ch/forum_thread.php?id=1222

and

http://lhcathome.cern.ch/old_news.php

As I said...it's all there (even a comment about how fast the wu's disappear when there are some...lol).

Live long and crunch!

ID: 6938 · Report as offensive     Reply Quote
bass4lhc

Send message
Joined: 28 Sep 04
Posts: 43
Credit: 249,962
RAC: 0
Message 6947 - Posted: 10 Apr 2005, 17:02:37 UTC - in response to Message 6932.  

> If so, you have 3 choices:
>
> 1) Be patient and wait. Your clients will eventually down the 4.66 sixtrack
> 2) Reset the project which will delete everything in your LHC folder and
> download new work under the 4.66
> 3) Move to another project if the admins and the project are not meeting your
> expectations. There's nothing keeping you here if you are as unhappy as you
> appear to be....
>
> If you've bothered to look through the older posts, you'll see that the admins
> do NOT consider the "zero credit" WU's a waste as it's still the science being
> processed regardless, and that SHOULD be the most important reason to crunch,
> NOT the credits granted or the glory or whatever else drives a person.
>
> Does this help clear things up for you?

no, if i do option 1 i can wait the rest of my life. the old wu's will never end. they start over and over again.
option 2 is possible but no garantee.
option 3 is senseless. i donate cpu time to science. i don't care about ratings or whatever.

you seem to miss the point, i am not talking about zero credit wu's.
i am talking about wu's that will never report anything because they restart at 0% after every switch between projects.

i simply report a problem or 2 and you advise me to leave the project.
not the way to get results in science i think.

also, you suggest, like thegasgiant and other people, i should read older messages on the board. in this case i did, without result.
i have however simply not the time to read all the boards of the projects i donate to. do you mean boinc is only for people who have so many spare time so they can read all these boards?

i donate cpu time to science. i can however still be critical about the projects i donate to. it's not that strange if you donate something, you/i as donator wants to know what happens with what you/i donate.
this means i can expect info from the projects i donate to. do i get this? no. in your words, i am unhappy and should leave.
i have a better solution, lhc (and other boinc projects) should give more information. they simply could recognise that donated cpu time is something of value.

btw you did not mention option 4: stop donating anything.

ID: 6947 · Report as offensive     Reply Quote
Profile bluumi

Send message
Joined: 27 Sep 04
Posts: 13
Credit: 155,619
RAC: 0
Message 6950 - Posted: 10 Apr 2005, 22:06:49 UTC - in response to Message 6947.  


> you seem to miss the point, i am not talking about zero credit wu's.
> i am talking about wu's that will never report anything because they restart
> at 0% after every switch between projects.

Yes, we all miss the point :-)
You didn't explain the problem good enough. I read the message as many others, that you talk about the 4.64 Zero-CPU units.. your msg was to short.

I never saw that lhc restart WU's at 0% after project switching.
ID: 6950 · Report as offensive     Reply Quote
Profile Logan5@SETI.USA
Avatar

Send message
Joined: 30 Sep 04
Posts: 112
Credit: 104,059
RAC: 0
Message 6951 - Posted: 10 Apr 2005, 22:14:12 UTC - in response to Message 6947.  


> i am talking about wu's that will never report anything because they restart
> at 0% after every switch between projects.

> also, you suggest, like thegasgiant and other people, i should read older
> messages on the board. in this case i did, without result.
> i have however simply not the time to read all the boards of the projects i
> donate to. do you mean boinc is only for people who have so many spare time so
> they can read all these boards?


I did some research for you and came up with the following older threads of interest that are similar to your problem as you state it above.

One thing you'll notice is that they ALL have older BOINC Core Clients and GUI's as well as Sixtrack Versions in common....You can reach your own conclusions as to if that is a factor in your situation or not.

IMHO: I'd recommend that you make sure that you get the LATEST BOINC Core Client From Berkeley and reset LHC until you get the 4.66 Sixtrack then monitor your progress as you crunch the WU's.

http://lhcathome.cern.ch/forum_thread.php?id=1218
http://lhcathome.cern.ch/forum_thread.php?id=1177

http://lhcathome.cern.ch/forum_thread.php?id=1198#6163 Somewhat Similar...

And finally my apologies to you if I sounded Critical of you before. There's many people here who like to complain to hear their voices, and I didn't look past the complaint to see the valid question behind it when I saw your post in the early morning hours.... :]




ID: 6951 · Report as offensive     Reply Quote
Profile Logan5@SETI.USA
Avatar

Send message
Joined: 30 Sep 04
Posts: 112
Credit: 104,059
RAC: 0
Message 6952 - Posted: 10 Apr 2005, 22:19:51 UTC
Last modified: 10 Apr 2005, 23:50:46 UTC

Also, said by Project Admin Markku Degerholm in this thread: http://lhcathome.cern.ch/forum_thread.php?id=1271

Posted: 5 Apr 2005 17:01:28 UTC

"At least zero CPU time and progress indicator problems have been resolved.

We continue testing of the clients until our physicists are ready to generate more work."

Markku Degerholm
LHC@home admin



I now after reading this, STRONGLY recommend that you take my option to upgrade to the latest Sixtrack and BOINC Core Client more seriously....

ID: 6952 · Report as offensive     Reply Quote
bass4lhc

Send message
Joined: 28 Sep 04
Posts: 43
Credit: 249,962
RAC: 0
Message 6954 - Posted: 10 Apr 2005, 23:58:26 UTC - in response to Message 6951.  


> And finally my apologies to you if I sounded Critical of you before. There's
> many people here who like to complain to hear their voices, and I didn't look
> past the complaint to see the valid question behind it when I saw your post in
> the early morning hours.... :]
>
thanks for your answer logan 5.
and i gracefully accept your apologie ;)

even with the most new versions of the software the problem stays.
in an other part of this board i found a solution.
i keep every project in memory. this makes a difference.

it is however a problem which lhc@home should fix.
i donate spare cpu time. i did not hand over my computers.

friendly greetings,
bas
ID: 6954 · Report as offensive     Reply Quote
Profile Bruno G. Olsen & ESEA @ greenh...
Avatar

Send message
Joined: 17 Sep 04
Posts: 52
Credit: 247,983
RAC: 0
Message 6955 - Posted: 11 Apr 2005, 1:47:45 UTC - in response to Message 6947.  

> i donate cpu time to science.

Well, actualy that is not quite the whole story. As you are crunching for this project right now, you donate cpu time for science AND a beta test of this project ;) So, we're actually here to help making this project ready to go


ID: 6955 · Report as offensive     Reply Quote
Profile Thierry Van Driessche
Avatar

Send message
Joined: 1 Sep 04
Posts: 157
Credit: 82,604
RAC: 0
Message 6956 - Posted: 11 Apr 2005, 8:22:19 UTC - in response to Message 6954.  

> even with the most new versions of the software the problem stays.
> in an other part of this board i found a solution.
> i keep every project in memory. this makes a difference.
>
> it is however a problem which lhc@home should fix.

Agree with this.

If somebody starts with LHC@Home they will use it in the default way concerning the preferences. AFAIK, the default is leave in memory = no.

Best greetings from Belgium
Thierry
ID: 6956 · Report as offensive     Reply Quote
Profile The Gas Giant

Send message
Joined: 2 Sep 04
Posts: 309
Credit: 715,258
RAC: 0
Message 6958 - Posted: 11 Apr 2005, 10:19:18 UTC
Last modified: 11 Apr 2005, 10:19:54 UTC

I just saw today, with 4.64 mind you, after BOINC was shutdown and restarted that the LHC %completion and cpu time was reset to zero (yes zip, nadda, etc). Since I have 1GB of RAM I normally leave applications in memory so I have not seen this isssue before. I think I lost about 6 to 10 hrs of processing time, which is not too much but painful non-the-less.

3.2GHz Prescott, HT on, 1GB PC3200 RAM, XP SP2, BOINC V4.27 and LHC V4.64.

Live long and crunch!

Paul
(S@H1 8888)
BOINC/SAH BETA
ID: 6958 · Report as offensive     Reply Quote
Profile sysfried

Send message
Joined: 27 Sep 04
Posts: 282
Credit: 1,415,417
RAC: 0
Message 6961 - Posted: 11 Apr 2005, 11:35:57 UTC - in response to Message 6958.  

> I just saw today, with 4.64 mind you, after BOINC was shutdown and restarted
> that the LHC %completion and cpu time was reset to zero (yes zip, nadda, etc).
> Since I have 1GB of RAM I normally leave applications in memory so I have not
> seen this isssue before. I think I lost about 6 to 10 hrs of processing time,
> which is not too much but painful non-the-less.
>
> 3.2GHz Prescott, HT on, 1GB PC3200 RAM, XP SP2, BOINC V4.27 and LHC V4.64.
>
> Live long and crunch!
>
>
I have to confirm that... I have a dual cpu system and had to reboot.

the 2 units that were reset to 0% completition are my last 4.64 wu's....using CC 4.25 and XP SP2.
they're running again now...
tada for now.
ID: 6961 · Report as offensive     Reply Quote
Profile Bruno G. Olsen & ESEA @ greenh...
Avatar

Send message
Joined: 17 Sep 04
Posts: 52
Credit: 247,983
RAC: 0
Message 7063 - Posted: 18 Apr 2005, 14:30:58 UTC - in response to Message 6956.  

> AFAIK, the default is leave in memory = no.

That's correct - a thing I always forget when joining another project - until I notice a missing "(left in memory)" at the messages tab. It was a recommended setting for cpdn - apparently a recommended setting for other projects as well, so in my view default should actually be "yes" rather than "no"


ID: 7063 · Report as offensive     Reply Quote
Profile [AF>HFR] Linflaas
Avatar

Send message
Joined: 2 Sep 04
Posts: 4
Credit: 45,763
RAC: 0
Message 7082 - Posted: 19 Apr 2005, 18:58:18 UTC
Last modified: 19 Apr 2005, 19:35:19 UTC

Hi all,

Same problem 4.64 always restarting at 0%.

Reseting project to test newer version.

PIV 3.06 HT
1Gb Ram
Boinc 4.25

Add:
4.66+boinc 4.27, working well on linux AMD 2000+ Fedora FC2 1Gb
Honi soit ki mal y pense !
<img src="http://www.boincstats.com/stats/banner.php?cpid=332dfe11170cde798d4a66eb658a5f60">
ID: 7082 · Report as offensive     Reply Quote
Profile [AF>HFR] Linflaas
Avatar

Send message
Joined: 2 Sep 04
Posts: 4
Credit: 45,763
RAC: 0
Message 7174 - Posted: 23 Apr 2005, 8:26:11 UTC

Hi,

After resetting :

Boinc 4.25 + LHC 4.67 working fine
Honi soit ki mal y pense !
<img src="http://www.boincstats.com/stats/banner.php?cpid=332dfe11170cde798d4a66eb658a5f60">
ID: 7174 · Report as offensive     Reply Quote

Message boards : Number crunching : donated time wasted


©2024 CERN