1)
Message boards :
Number crunching :
sulphur model constantly slows down
(Message 20773)
Posted 26 Feb 2006 by mahdia Post: Hello Honza and Les Bayliss! Thank You alot for Your advices. Whatever it came from, the slowdown is (nearly) away. I only cleared up my hard disk, no other change was made. The last trickle (occured some minutes ago) showed me a sec/TS value of ca. 4.57, relative to the last trickle, which has 8.01 sec/TS. I made some calculations, which show me an average 0.03458 sec/TS increase per trickle (total), the average increase relative to the pref trickle is 0.0688 sec/TS. I can\'t imagine, that a defragmented disk make such a great difference. How many bytes are written to disk during checkpointing ? Regards |
2)
Message boards :
Number crunching :
sulphur model constantly slows down
(Message 20761)
Posted 26 Feb 2006 by mahdia Post: It shouldn\'t really slow that much. I have an P4 550 processor (prescott), working on 3.4 GHz. As I understood the infos from the Intel WebSite, there is no such way of thermal protection mechanism for this particular processor stepping (F-3-4-D0) (or should I be missed something). The cpu temperature is in the range from 59-62 degrees Celsius while working on CPDN model, I dont\'t know, if this is to hot. Whatever, I will watch the trickles after my disk clean action is completed and if this will not help, I have to clean the box from dust. Regards |
3)
Message boards :
Number crunching :
sulphur model constantly slows down
(Message 20760)
Posted 26 Feb 2006 by mahdia Post: do some maintanance like disk defragmentation, spyware/adware check etc. and resume BOINC. Oh my god, it looks like a great \"tohuwabohu\" on the hard disk, spyware also found. Mmh, I thought, my Norton Antivirus Package already does this spyware check :-( Ok, I will clear up the hard disk and watch the trickle time (hopefully) speeds up. Thanx. |
4)
Message boards :
Number crunching :
sulphur model constantly slows down
(Message 20757)
Posted 26 Feb 2006 by mahdia Post: Hello ! My (P4-HT) machine computes a sulphur model. It started with 4.09 s/TS on begin of phase 1 and have now reached 6.03 s/TS in the middle of phase 3 (see here: http://climateapps2.oucs.ox.ac.uk/cpdnboinc/trickle.php?resultid=1347636). Question: Is this a normal behaviour of a sulphur model or should I make myself concerns about this machine? Regards |
5)
Message boards :
Number crunching :
server state means ??
(Message 18272)
Posted 16 Dec 2005 by mahdia Post: Hi, Honza ! I aggree with you, that the CPDN behavoir is ok, but I wondered about the over state on a work unit, that hasn\'t reached dead line. I only want to be sure not to crunch for the paper-basket. Regards, Frank |
6)
Message boards :
Number crunching :
server state means ??
(Message 18266)
Posted 16 Dec 2005 by mahdia Post: I have the same question too. One of my results (1347636) shows the \"over\" state, but this result is still crunching, trickling and getting credits. |
7)
Questions and Answers :
Windows :
return value of -182
(Message 16762)
Posted 25 Oct 2005 by mahdia Post: Hello ! After updateing to version 5.2.2 I constantly get the following error mesage: 25.10.2005 08:48:42|climateprediction.net|Scheduler request to http://climateapps2.oucs.ox.ac.uk/cpdnboinc_cgi/cgi failed with a return value of -182 There is the same message with other projects. What should I do ?` Regards, Frank |
8)
Questions and Answers :
Windows :
moving boinc folder from c: to d:
(Message 15576)
Posted 31 Aug 2005 by mahdia Post: How to do this best without deinstalling ? Which files are to update after a move ? BOINC runs as a service. Regards, Frank |
9)
Questions and Answers :
Windows :
model crash after stop and start boinc as a service
(Message 7260)
Posted 12 Jan 2005 by mahdia Post: Thanx Les ! I try to remember the problem everytime shutting down and restarting my computers, but I cannot promise much ;-) I thought it was an cpdn problem because there is always a cpdn WU affected, never seen this behaviour an seti or lhc WUs. Nevertheless, I had walked a bit through the BOINC code and I finally think, it's a timing problem how long boinc client resp. windows service control manager waits that the application completes the stop-request until it kills the process. BTW: the problem never arises when running boinc as normal application and not as a service. So it seems a problem handling windows service control manager request. I will try to work out some fix, because I myself have developed a lot of windows services. Regards, Frank |
10)
Questions and Answers :
Windows :
model crash after stop and start boinc as a service
(Message 7253)
Posted 11 Jan 2005 by mahdia Post: Hello ! It is obviously, that cpdn scientific application has problems to correctly handle the stop and (re)start when running boinc as a service. Although windows service control manager reports no problems stopping and starting the boinc service, I can reproduce an error resp. model crash stopping the service during cpdn model is in initialisation process after (re)start. Perhaps the developers can take a look at this to avoid model crashes (and wasted time), I will try my best to work around this error. Regards, Frank |
11)
Message boards :
Number crunching :
great differences in time to completion but (nearly) same benchmark results
(Message 6795)
Posted 9 Dec 2004 by mahdia Post: Hello, Carl ! Are you already in Potsdam ? > completion times. CPDN also does not use the "official" BOINC benchmarks, > credits are based on the trickle points (something like 90 per 10,802 timestep > trickle I believe)? I really don't care about credits, simply try to make the runs faster. Greetings from Berlin, Frank |
12)
Message boards :
Number crunching :
great differences in time to completion but (nearly) same benchmark results
(Message 6730)
Posted 8 Dec 2004 by mahdia Post: Hello, Geophi ! > Hard to know for sure without knowing more about your systems. What are the > PCs you are referring to? Thanks for your reply. Both Intel Pentium 3 processors, one with 800 MHz, the other 1 GHz. But I think, the processor isn't the reason. There are complete differences in mainboard, memory etc. For CPDN I looked at the time/model time step: 3.4 s for pc1, 7.9 s for the other. So it seems to me, that I have to wait a long time to complete the model. Mmh... Greetings, Frank > > The BOINC benchmarks aren't really anything like the CPDN model and are not an > indicator of performance for CPDN. If the processors are two of the same type > (P4, Celeron, AthlonXP, Athlon64) and same speed, then the difference could be > due to motherboard/memory differences. CPDN performance is highly dependent > on memory bandwidth and latency. Anything that increases memory bandwidth > and/or decreases memory latency while still being stable, will result in > better performance. > > Otherwise, if performance ratings are approximately the same, and one > processor is an Intel and the other AMD, the Intel will likely be faster > because of optimizations used in compiling CPDN. > |
13)
Message boards :
Number crunching :
great differences in time to completion but (nearly) same benchmark results
(Message 6716)
Posted 7 Dec 2004 by mahdia Post: Hello, BOINC fellows ! I try to understand the following: I have to machines, which produce nearly the same BOINC benchmarks (715/740 whetstones, 1730/1780 drystones), but computing a CPDN model takes on the first ca. 1600 hours (uff, long time, I know), on the other (that with a little better benchmark results) expects to take ca. 900 hours. Why are the great differences ? And what can I do to make it faster (please no advices to buy a faster machine ;-)) Grettings, Frank |
©2024 climateprediction.net