climateprediction.net home page
Posts by Jean-David Beyer

Posts by Jean-David Beyer

1) Message boards : climateprediction.net Science : Climate change in the News (Message 62091)
Posted 18 days ago by Jean-David Beyer
Post:
+18C in Antarctica today. (11C higher than Cambridge!)


Since there is no such thing as climate change, there must be severe Orgone Fields sweeping over Antarctica to cause all their thermometers to read so much higher than the real temperatures.
2) Questions and Answers : Unix/Linux : *** Running 32bit CPDN from 64bit Linux - Discussion *** (Message 62076)
Posted 23 days ago by Jean-David Beyer
Post:
Ubuntu now includes 32-bit libraries by default.
HadAM4 now is 64-bit


Not on my 64-bit machine. All seem to be 32-bit.

$ file hadam4_8.09_i686-pc-linux-gnu hadam4_8.52_i686-pc-linux-gnu hadam4_um_8.09_i686-pc-linux-gnu hadam4_um_8.52_i686-pc-linux-gnu hadcm3s_8.36_i686-pc-linux-gnu hadcm3s_um_8.34_i686-pc-linux-gnu hadcm3s_um_8.36_i686-pc-linux-gnu

hadam4_8.09_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not stripped
hadam4_8.52_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not stripped
hadam4_um_8.09_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not stripped
hadam4_um_8.52_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not stripped
hadcm3s_8.36_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not stripped
hadcm3s_um_8.34_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not stripped
hadcm3s_um_8.36_i686-pc-linux-gnu: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.18, not stripped
3) Message boards : Number crunching : Scheduler request too recent (Message 61946)
Posted 10 Jan 2020 by Jean-David Beyer
Post:
Which doesn't take into account that most people joining over the past couple of years seem to have no particular interest in the climate models, and load up with lots of work from other projects as well.


I joined quite a while ago. I am particularly interested in climate models so my priorities are 50% climateprediction, 25% WCG, 13% Rosetta, and 12% Seti@home. When I can get all the work units I wanted, this ended up with three processors running CPDN, and the others taking their turns on the other processor.

These days there are large gaps in work-unit availability, so now I have Rosetta and Seti@home set to Won't Get New Tasks. I allow new tasks in CPDN until I get four work units, and then turn off getting new ones. I leave WCG on, but I accept only one work unit at a time on two of their jobs, and two at a time on one other. As a result, I seldom get any.

Right now, my deficit on CPDN work is so great that I am usually getting four work units at a time. (Right now I am running one N144 work unit and three N216 work units.) But when this does not happen, I get a work unit or two from WCG that usually go through in less than a day. If I run out of work, I enable either seti@home or rosetta just long enough to get some work units (usually 6 or 8 come through) and they grab the processors. By the time they are done, I have received some more CPDN work. If work units were available more of the time, I hope to find a way to just let them all run all the time without my fiddling around like this. I.e., like in the old days.
4) Message boards : Number crunching : UK Met Office HadAM4 at N216 resolution (Message 61865)
Posted 31 Dec 2019 by Jean-David Beyer
Post:
This one failed for me due to negative pressure.
I do have the required 32-bit libraries as evidenced by other 144 work have completed successfully, as have the 216 ones.

Task 21866633
Name hadam4_a06t_209110_6_856_011961960_3
Workunit 11961960

My failed one says.

Workunit 11961960
name hadam4_a06t_209110_6_856_011961960
application UK Met Office HadAM4 at N144 resolution
created 9 Dec 2019, 1:30:09 UTC
minimum quorum 1
initial replication 1
max # of error/total/success tasks 5, 5, 1
errors Too many total results

No need to do anything about this, I suppose: just another data point.
5) Message boards : Number crunching : What ever it was modeling it did it fast. (Message 61810)
Posted 25 Dec 2019 by Jean-David Beyer
Post:
Seems there sadly are not many Linux crunchers out there, or at least connected to this project in comparison to others.


I have been signed up for CPDN for ages, running only Linux. If memory serves, I went about a whole year without getting any CPDN work.
Presently, I am running one hadam4h and two hadam4 work units. Also one WCG work unit.
My normal intention is to run 50% CPDN, 25% WCG, and 12.5% each of Seti@home and rosetta@home. I enable these last two only when I have idle processor cores and cannot get either CPDN or WCG work units. When the latest hadam4 and hadam4h Linux units started coming out, I was running 100% CPDN for a while.

I do not know if you would count me as a Linux cruncher during the year when I crunched no CPDN work units, because I did not get any.
6) Message boards : climateprediction.net Science : Climate change in the News (Message 61706)
Posted 18 Dec 2019 by Jean-David Beyer
Post:
I wonder why a study was required for this. When I was in high school in the mid 1950s, I learned that the solubility of gasses in water went down with increasing temperature. I cannot imagine this has changed since then. The study may reveal to the ignorant just how bad the situation already is and that, unless we make a dramatic change in energy use, what the future is likely to be.
7) Message boards : climateprediction.net Science : Climate change in the News (Message 61635)
Posted 2 Dec 2019 by Jean-David Beyer
Post:
If they set it up as BOINC work units, the cloud of boinc users is already here and ready to go, no?
8) Message boards : Number crunching : New work Discussion (Message 61625)
Posted 30 Nov 2019 by Jean-David Beyer
Post:
All my N216s were 8.53


Oops! Typo. I meant 8.52.
9) Message boards : Number crunching : Why did I get ZERO credit for these tasks? (Message 61619)
Posted 27 Nov 2019 by Jean-David Beyer
Post:
And eventually all of this will be put into files for the external stats sites to collect later.


Some external sites refer to this reasonably promptly.

But this site:

https://boinc.netsoft-online.com/e107_plugins/boinc/get_user.php?cpid=549ef8ba1b5181207e1ad3a2ba626d83&html=1

does not do it hardly ever anymore. It thinks I have 3,075,732 credits while your site says I have 3,239,868; they are about 164,000 credits behind.
10) Message boards : Number crunching : New work Discussion (Message 61591)
Posted 21 Nov 2019 by Jean-David Beyer
Post:
My N144s that worked were all 809. the one 808 I got crashed.
All my N216s were 8.53; the first one was from 17 Oct 2019, 6:34:33 UTC
that completed 9 Nov 2019, 14:18:39 UTC that shows how slow (or fast) my 1.8 GHz Xeon processor it.

They certainly checkpoint more frequently than the, send trickles. The most recent checkpoint was about an hour ago, but that does not prove much.
11) Message boards : Number crunching : Validation pending for 9 years... (Message 61584)
Posted 20 Nov 2019 by Jean-David Beyer
Post:
I have 77 still pending for validation, but I am not worried either. They start sometime in 2010 and end in December 2015.
I have 157 valid ones. I have this much credit: 3,212,752

I wonder what it all means.
12) Message boards : Number crunching : Validation pending for 9 years... (Message 61578)
Posted 19 Nov 2019 by Jean-David Beyer
Post:
5 GB per core?! I do run a mixed number of memory light and memory heavy workloads, but so far the biggest one used "only" 2GB per core.


My biggest ones (hadam4_um_8.52_i686-pc-linux-gnu N216) each use 1385 Megabytes of virtual memory, but actually resident in RAM is only 1.3 Gigabytes.

Are you perhaps referring to OpenIFS models? I have not seen any of those yet.
13) Message boards : climateprediction.net Science : Misconfigured Machine? (Message 61576)
Posted 19 Nov 2019 by Jean-David Beyer
Post:
There is some discussion about this at the moment and whether sending emails to owners is GDPR compliant.


If I remember correctly, when signing up for this group, I checked a box saying it was OK to send me e-mails from the project. If this is correct, it seems to me the project could send people who did this e-mails.
14) Message boards : climateprediction.net Science : Misconfigured Machine? (Message 61571)
Posted 18 Nov 2019 by Jean-David Beyer
Post:
Looking at this thread, should a Wall of Shame be established to list all the machines that crash due to missing 32-bit libraries?
If so, could it be populated automatically?
Could e-mails be sent automatically to the offenders so they could fix the problems, or could they be disconnected until they get their house in order?
15) Message boards : Number crunching : UK Met Office HadAM4 at N144 resolution (Message 61568)
Posted 18 Nov 2019 by Jean-David Beyer
Post:
Apparently the N144 should have lower memory requirements than the N216, because of the lower resolution.


They do. On my Linux 64-bit machine with 16 GBytes RAM, they each required about 4% of my RAM*; the N216 ones each require between 8.5% and 8.6% (1.3 GBytes).
_____
* I just realized, I do not remember if I was running the N144 tasks with 8 GBytes RAM or 16, but if it was 8, the difference in size would be even greater.
16) Message boards : Number crunching : Validation pending for 9 years... (Message 61560)
Posted 17 Nov 2019 by Jean-David Beyer
Post:
I am also curious if 32-bit library requirement for Linux is still current?


It is still current.
17) Message boards : Number crunching : UK Met Office HadAM4 at N144 resolution (Message 61554)
Posted 16 Nov 2019 by Jean-David Beyer
Post:
I have a horrible feeling that I'm going to have to do a lot fiddling with the pref settings to fend off N216 downloads while I try and get some N244.


Not right now; I assume you mean N144. There seem to be no N216 tasks at the moment.

UK Met Office HadAM4 at N144 resolution 1808 1640 126.25 (64.58 - 329.88) 12
UK Met Office HadAM4 at N216 resolution 0 3862 392.06 (128.25 - 655.58) 29
18) Message boards : Number crunching : UK Met Office HadAM4 at N144 resolution (Message 61551)
Posted 16 Nov 2019 by Jean-David Beyer
Post:
In general, hyper-threading helps as Dave said. So I would try it.

But on the larger ones (N216), it hurt. That was not because there was anything wrong with HT itself, but you were running out of cache memory (on the CPU) with so many work units running at once.


I used to have a machine with two 32-bit 3.06 GHz Xeon processors that could be hyperthreaded, so it appeared as having 4 processors. I do not recall how much cache those two chips had. I used to run Seti@home, climateprediction, rosetta, and WCG. I tended to run three climatepredictions and one other. Now hyperthreading four processors (i.e., with hyperthreading turned on) would turn out more work than two, but not twice as much. So each task proceeded more slowly that way, but the total tasks per a day was more.

My current (slow 1.800 GHz) processor has four 64-bit cores, but 10240K of cache. I cannot hyperthread them. I run Linux. I am currently running four N216 processes and they are getting 92%, 92%, 96%, and 97% of a processor. It is taking Average (sec/TS) 53.6570, but it runs so slowly that I do not wish to stop two of the processes to see if this would improve the cache hit ratio. It seems to take almost three weeks for me to do an N216 task, and that N144s ran faster. Average (sec/TS) 25.8696, taking me about two weeks.

I suspect that since these processes are in a big loop, that they are probably running the same code, so the instructions in the cache may only be in there once (once the program gets started, say after a few hours). So cache misses may be less of a problem that at first appears. This would not apply if one were running different applications (such as WCG, or even hadcm3s). Of course the data will be different, and that will increase the probability of a cache miss.
19) Message boards : Number crunching : no new work units (Message 61531)
Posted 14 Nov 2019 by Jean-David Beyer
Post:
the new models seem to need around 4 Megs of L3 cache, or they slow waaaaaaay down.


Hurray! My slow (1.8GHz) 4-core 64-bit processor has 10240 Megs of L3 cache.
And now I have 16GBytes of DDR3 registered RAM in eight 2 GByte modules.
20) Message boards : Number crunching : no new work units (Message 61525)
Posted 11 Nov 2019 by Jean-David Beyer
Post:
Right now you only have two kinds of workunits available, one that takes an average of 147 hours and one that takes an average of 337 hours that's a huge difference especially if your pc is older and doesn't run in the 3 or 4 ghz range for speed.


On my slow machine, 1256552, the N144 work units are taking about 16 days each, and the N216 work units are taking about 22 days. Machine running 24/7, and not doing much else except web browsing and e-mail. I am currently running four N216 work units.

CPU type 	GenuineIntel Intel(R) Xeon(R) CPU E5-2603 0 @ 1.80GHz [Family 6 Model 45 Stepping 7]
Number of processors 	4
Operating System 	Linux 2.6.32-754.23.1.el6.x86_64
BOINC version 	        7.2.33
Memory 	       15.5 GB
Cache 	      10240 KB
Swap space     3.91 GB
Total disk space 	117.21 GB
Free Disk Space 	 97.36 GB
Measured floating point speed 	1.27 billion ops/sec
Measured integer speed 	        3.53 billion ops/sec
Average upload rate 	    1394.09 KB/sec
Average download rate 	   10519.69 KB/sec
Average turnaround time 	18.94 days


Next 20

©2020 climateprediction.net