1)
Message boards :
Number crunching :
Odd validation (Linux Client sometimes 10x faster than Win32 ?)
(Message 6454)
Posted 6 Mar 2005 by Jim Harris Post: > That leaves me with 2 possibilities : > - Validator screwed up OR > - Linux Client is sometimes 10x faster on some WorkUnits > > I wouldn't mind getting upto 18x more Credits Granted than Claimed (my Linux > BOINC Cores are fully optimized, therefor they claim a realistic number), but > that "speed advantage" seems waay to big to be "normal" by any means. > I believe you experienced a 3rd possibility that other users are getting hit with is that cpu time is being reset to 0 and then counts up till workunit completes. Also there is text in your stderr out: -core_client_version>4.19-/core_client_version> -stderr_txt> boinc_init() v4.61 boinc_init() v4.61 boinc_init() v4.61 boinc_init() v4.61 boinc_init() v4.61 boinc_init() v4.61 boinc_init() v4.61 boinc_init() v4.61 boinc_finish(0) -/stderr_txt> Other users do not have this text. I also have an example with a 1,000,000 turn boince workunit: http://lhcathome.cern.ch/workunit.php?wuid=76170 One user reported with a time of 845.61 seconds and claimed credit of 1.28, I had a time of 33,748.80 seconds and claimed credit of 99.67. Granted credit was 1.28 :( Interestingly this user also has linux os and stderr out text the same as yours. |
2)
Message boards :
Number crunching :
Disk access every second
(Message 5913)
Posted 22 Feb 2005 by Jim Harris Post: > Yes, and how many times is very frequently? What have you set your preferences > on the site to write to disk at? Do you use these same preferences for other > projects? > > There are projects which ignore the standard BOINC setting of 60 seconds, and > write to disk earlier or only once every 30 minutes. > > My preference is set to write to disk every 10 seconds. > I made sure I am spyware free, virus free, worm free. I don't index any NTFS > drives when part of my system is idle. > > So how can you be sure that when you see the light on your harddrive blink, > that it is writing part of LHC to the harddrive? > Get a grip on yourself. I posted to confirm PeteHow was correct. And that there was not a spyware, virus issue or anything else involved. I can be sure because I used 2 programs to verify that it was LHC's application writing to disk. I never stated it was a problem. @Ageless - don't dismiss someone's post so easily. |
3)
Message boards :
Number crunching :
Disk access every second
(Message 5888)
Posted 22 Feb 2005 by Jim Harris Post: > Yet try a unit of 100,000 of 1,000,000 cycles and see if it still writes to > disk every second of every minute. > Currently I'm running a 100,000 turn wu. It's writing to 30 fort.xx files very frequently. |
4)
Message boards :
Number crunching :
Disk access every second
(Message 5882)
Posted 22 Feb 2005 by Jim Harris Post: PeteHow is correct. The LHC application is writing to a large number of files in the slots directory very frequently. The files are named fort.xx where xx is a number. |
5)
Message boards :
Number crunching :
Hooray for the new workunits
(Message 3313)
Posted 7 Oct 2004 by Jim Harris Post: |
6)
Message boards :
Number crunching :
Forums slow & database connection errors
(Message 3100)
Posted 2 Oct 2004 by Jim Harris Post: |
7)
Message boards :
Number crunching :
It appears that cheating has arrived for BOINC & LHC
(Message 3006)
Posted 1 Oct 2004 by Jim Harris Post: |
8)
Message boards :
Number crunching :
sience crunching or credits crunching ,what you like ?
(Message 2991)
Posted 1 Oct 2004 by Jim Harris Post: |
9)
Message boards :
Number crunching :
CPU
(Message 2985)
Posted 1 Oct 2004 by Jim Harris Post: |
10)
Message boards :
Number crunching :
what is tunescan?
(Message 2837)
Posted 30 Sep 2004 by Jim Harris Post: |
11)
Message boards :
Number crunching :
what is tunescan?
(Message 2796)
Posted 29 Sep 2004 by Jim Harris Post: |
©2024 CERN