climateprediction.net home page
Posts by geophi

Posts by geophi

1) Message boards : Number crunching : No trickles on webpage (Message 61971)
Posted 5 days ago by Profile geophi
Post:
These N216 models seem to have semi-regularly spaced time-steps that take more than 4x CPU time than most of the other other time-steps. I think it is a little bit like the "key frames" used to encode video. It does make sense if you consider that in the models every "cell" (temp, air pressure, humidity, and a lot more than I think off of the top of my head) in the 3D matrix will influence it's neighboring cells (and those cells must influence their neighbors and so on until you reach the edges of the zone being modeled) at each time-step. If this is true then sooner or later a model will reach a "large change" state where one or more cells change in a way that critically influences the entire model.

Those are the timesteps that include the radiation calculations. Most of the models do that calculation every 3 hours.

The hadam4/hadam4h models are global models with no regional component. The Weather At Home (WAH) models that run on Windows have both global and regional components, where the global model (at a lower resolution) is run for one day, then the regional model grid is run for that day, with boundary conditions from the global model, then on to the next day, etc.
2) Message boards : Number crunching : No trickles on webpage (Message 61969)
Posted 6 days ago by Profile geophi
Post:
If interested, you can check the average sec/TS by going into the

"boinc cliient directory"/projects/climateprediction.net/"name of model task directory"

and tailing the stdout_mon.txt file

tail -f stdout_mon.txt
3) Questions and Answers : Unix/Linux : About hadam4_a0tt_209410_6_856_011962788_0 (Message 61935)
Posted 8 days ago by Profile geophi
Post:
Any tasks that are on a PC when a project reset is requested, will continue to be listed as "In Progress" until they time out. So that task won't be reissued for quite awhile. This happens occasionally and the project can deal with it, so it's not something to be too concerned about.
4) Message boards : Number crunching : No trickles on webpage (Message 61929)
Posted 8 days ago by Profile geophi
Post:
@Bonsai911

How far along (what is the progress) for the tasks that aren't trickling? You have a whole bunch of tasks on that PC that are trickling just fine and have been all along. Usually when someone reports missing trickles, it's because a server reboot occurred and the trickle storage task didn't get restarted. The missing trickles here don't look to be a server side problem.
5) Questions and Answers : Unix/Linux : Dell T610 12 cores 96 GB RAM Linux Ubuntu no GUI, how to start computing (Message 61920)
Posted 10 days ago by Profile geophi
Post:
Not sure where you are at on this. Did you run this command for 32bit compatibility?

sudo apt-get install lib32ncurses5 lib32z1 lib32stdc++-6-dev

If you installed boinc from the Ubuntu repository, I think the boinc service starts automatically when it boots up. I see in the listing in your message that the URL is http://climateprediction.net
I'm not sure if this is the problem, but the URL to attach to should now be https://climateprediction.net/

Even if all that is okay, right now there is no work for Linux, although there should be some more later this week.
6) Message boards : Number crunching : Dual Opteron 6128, Dual Xeon E5-2670, both, or neither? (Message 61896)
Posted 13 days ago by Profile geophi
Post:
The Opteron would be of Phenom II performance, (circa 2009/2010) but limited to 2 GHz in speed, and power hungry. I'm thinking it's not a good fit for the current models.

The Xeon on the other hand is of Sandy Bridge era performance, has much more cache, higher base CPU speed, turbo boost possible, quad channel memory capability, faster memory capability, etc. Obviously wouldn't want to run on the Hyperthreads, but otherwise a capable processor/system.
7) Message boards : Number crunching : Scheduler request too recent (Message 61867)
Posted 16 days ago by Profile geophi
Post:
So, I'm getting this 3600+ second wait even though NO work has been sent to the client. I'm wondering how many hours it will sit idling, waiting for work that may never come?

According to the server status page, there is no work for Windows computers at this time (WAH2 models). https://www.cpdn.org/server_status.php

These come in batches and we as moderators seldom know ahead of time when a batch will be released. With a lot of Windows computers attached, the tasks in the queue don't take long to be snapped up once the batch is out there. Sorry I can't bring better news at this time.
8) Message boards : Number crunching : UK Met Office HadAM4 at N144 resolution (Message 61846)
Posted 21 days ago by Profile geophi
Post:
I do not mean to be a pain but I have to ask if that means a.) "DON'T schedule more threads than you have physical cores," b.) "DON'T have enabled Hyper-Threading in your BIOS," or c.) "DON'T have ANY symmetric multi-threading from ANY vendor enabled in your BIOS."

I would choose a there.
9) Questions and Answers : Windows : Unfinished tasks (Message 61843)
Posted 21 days ago by Profile geophi
Post:
It is possible to change the automatic update installations. Search for it on the net, but you will have to be confident about making changes to ther registry.

Windows 10 version 1903 gives you more ability to delay when the update/restart will happen. These are set under the Settings app. Previous to that version, there was very limited control for Windows 10 Home users.
10) Message boards : Number crunching : UK Met Office HadAM4 at N144 resolution (Message 61839)
Posted 21 days ago by Profile geophi
Post:
I was interested in what effect memory speed/bandwidth might have on the N144 models. While this is not a complete, or ideal test, I ran two N144 models on my 4790K and changed the memory speed to calculate the difference in sec/TS based on the changes. The PC was running at 4.4 GHz through all tests and had DDR3 2400 capable memory in it.

1600 MHz 8.23 sec/TS
1866 MHz 7.80 sec/TS (5.2% faster than 1600 MHz)
2133 MHz 7.58 sec/TS (7.9% faster than 1600 MHz)
2400 MHz 7.45 sec/TS (9.5% faster than 1600 MHz)

How this would change running 4 at a time, or running 2 or more N216 models, is unknown at this time. I was just curious.
11) Message boards : Number crunching : Computation error, newly added project (Message 61836)
Posted 22 days ago by Profile geophi
Post:
I run some of my Linux boxes on UTC/GMT time, so that may be why I remember that.
12) Message boards : Number crunching : Computation error, newly added project (Message 61830)
Posted 22 days ago by Profile geophi
Post:
geophi, "cannot find the device/drive specified" - what are you referring to?

Your Windows computer's tasks are here https://www.cpdn.org/results.php?hostid=1496463

If you click on the individual task number, you'll see a section labeled stderr where some additional errors are written. The problem up near the top of that listing is what resulted in the task error "The system cannot find the drive specified."
13) Message boards : Number crunching : Computation error, newly added project (Message 61825)
Posted 22 days ago by Profile geophi
Post:
It's possible that the "day" will reset at 00 GMT, so it may ask for work after that.

The Windows errors are odd, with the "cannot find the device/drive specified" problem. I've had these types of errors occasionally, but not in bunches like that. It may have to do with the system trying to do too much disk reading and writing simultaneously, but that's just a hunch. If those types of errors continue frequently, I'm not sure what the solution would be.
14) Message boards : Number crunching : Computation error, newly added project (Message 61821)
Posted 22 days ago by Profile geophi
Post:
On your 2600X, you don't have the 32bit libraries loaded that climateprediction.net needs. Sticky at the top of the Linux forum.

https://www.cpdn.org/forum_thread.php?id=8008&postid=59939

Since you're using 19.2, it's based off of Ubuntu 18.04 so use that command in the sticky to get the needed libraries.

Edit...also, you only have 8 GB of RAM on the 2600X with 12 cores that boinc sees. This is problematic in terms of memory usage if running on all cores, and I would suggest limiting the number of CPUs used by boinc to at most 6 for the hadam4 N144 models, or at most 4 if you get the hadam4h N216 models.
15) Message boards : Number crunching : Scheduler request too recent (Message 61819)
Posted 22 days ago by Profile geophi
Post:
It appears to be set at an hour plus a few seconds

Right. It will request work only after 1 hour of no communication with the project. So clicking the Update button in the Projects tab of boinc manager with climateprediction.net selected is counterproductive for requesting new work, unless it's been over an hour since the last scheduler request.
16) Message boards : Number crunching : Why no new projects to run? (Message 61783)
Posted 24 days ago by Profile geophi
Post:
WAH2 models are Windows only. HADCM3N (Coupled model full resolution) also, although they haven't been issued for years.
17) Message boards : Number crunching : UK Met Office HadAM4 at N144 resolution (Message 61766)
Posted 26 days ago by Profile geophi
Post:
Are the n216's different as to when they send a trickle? And I'm assuming WAH under Windows is likewise different, so will now keep a close eye on that too.

The n216's also trickle each model month, just like the n144's. However, it takes well over a day of computing time (or more depending on computer speed and number of models running on a PC) to finish a month and produce a trickle.
18) Message boards : Cafe CPDN : Australia's bush fires (Message 61765)
Posted 26 days ago by Profile geophi
Post:
Is the smoke bad where you live Les?
19) Message boards : Number crunching : New work (Message 61755)
Posted 27 days ago by Profile geophi
Post:
And it's a little more complicated than the back of the envelope calculation. So many of the tasks in progress, really aren't in progress. With deadlines nearly a year, tasks having been abandoned in certain ways on some hosts, that for whatever reason, aren't crunching any more, are still seen as in progress. How many haven't trickled for 2 months or more, I don't know, but I'm sure it's a lot. The deadline should really be shortened to something far less than a year (1-3 months depending on the model type?) so that that number is more representative of the number in progress, and the tasks get re-issued quicker if they time out on a host no longer crunching. Just my opinion...
20) Message boards : Number crunching : New work (Message 61746)
Posted 27 days ago by Profile geophi
Post:
They must be dumping more into that SAM50 batch as they're are still quite a lot in the queue now.


Next 20

©2020 climateprediction.net