climateprediction.net home page
Posts by Jim1348

Posts by Jim1348

1) Message boards : Number crunching : Updated BOINC Clients 7.16.11 - Windows 64-bit and Mac OS X (64-bit Intel) (Message 62721)
Posted 12 Sep 2020 by Jim1348
Post:
The various academic institutions around the world that work on climate science do much of their work probably most of it on supercomputers. There are specific things they send work via Oxford to crunchers for.

Another issue is the type of work it will be. In the most basic form, the question is Windows or Linux?
The Windows work is fairly uniform in its system requirement, but Linux is another matter. Memory and cache play a large role in how many you can run.
So you really have to plan for the worst-case, and set your machines up accordingly if you want to get the first ones to work properly.
(We can't select of course.)
2) Questions and Answers : Unix/Linux : No more new tasks (Message 62681)
Posted 6 Sep 2020 by Jim1348
Post:
In the meantime Yanngo, you should find out why all but one task sent to your computer has crashed.

As a wild guess:
Memory 2.83 GB

Also, there are a whole lot of suspensions. If you are going to suspend, at least have enough memory for "leave it in memory" to work.
3) Message boards : Number crunching : Big models (Message 62647)
Posted 8 Aug 2020 by Jim1348
Post:
It's just that, as a species, I think we a dumber than a bag of hammers.

We haven't exactly distinguished ourselves.
I do a lot of COVID-19 work, but the anti-virals that will save us near-term are already under test now, and should be available by the end of the year.
The computer studies I think are more relevant for the next-generation viruses. We need to start work now.
4) Message boards : Number crunching : Big models (Message 62644)
Posted 7 Aug 2020 by Jim1348
Post:
These are the hadam4h N216 models that have been the main batches we've been running lately on Linux.

Thanks. I limit them to two (or maybe four) per machine for maximum efficiency.
The old ones have run well on a Ryzen 3600 (virtual cores) or i7-9700 (full cores) with that, and I expect the new ones will too.
5) Message boards : Number crunching : Big models (Message 62641)
Posted 7 Aug 2020 by Jim1348
Post:
OK, but it seems a little small. I will be taking a Ryzen 3600 out of summer lockdown in a month or so, and with a 20 Mpbs (or 2.5 MBps) upload speed, I won't have that much to do.
But give it your best shot, please.
6) Message boards : climateprediction.net Science : AFlame PROJECT (Message 62617)
Posted 19 Jul 2020 by Jim1348
Post:
What I would really like to see though is all Linux tasks moving to openIFS and an end to tasks being trashed by users who don't read the forums and don't install the required 32bit libs.

I wonder if openIFS is merely dead, or dead as a door knob?
7) Message boards : Number crunching : Please Explain Scheduling Priority (Message 62592)
Posted 19 Jun 2020 by Jim1348
Post:
I'm trying to figure out why it only looks for tasks once a day and never gets any.

That is because there aren't any. Check the Server status page. You just got a resend on the one machine.
Once a day is normal I suppose. Sometimes it is less, an hour or so, when tasks are available.

The "scheduling priority" is a negative number. 0.00 means that it is the highest priority project and should get the next one.
A more negative number (-0.10) means that it is lower priority project, and won't get one if the buffer is full (or run it) until it moves up the order. It is a strange system.
8) Message boards : Number crunching : New work Discussion (Message 62561)
Posted 8 Jun 2020 by Jim1348
Post:
You can of course control the size of your buffer. Setting it to the default of 0.1+0.5 days will get you fewer work units than setting it to 10 days, for example. But that is rather crude, and sometimes they won't send you anything at all.

Unfortunately, the BOINC scheduler is not aware of the "app_config.xml" file, so if you set it to run only two work units, it will still download as if you were running on all 12 cores. So until BOINC gets smarter (a long process), the best way would be for CPDN to provide you a user-selectable limit of how many work units you want downloaded at a time. Several projects do that now: WCG, LHC and Cosmology that I can think of at the moment.

But CPDN has been going in the direction of fewer options. They need to reverse course.
9) Message boards : Number crunching : Uploading files fails (Message 62533)
Posted 28 May 2020 by Jim1348
Post:
A couple of dozen zips have just uploaded for me; only two left.
10) Message boards : Number crunching : Uploading files fails (Message 62524)
Posted 27 May 2020 by Jim1348
Post:
It is fun commiserating when you can't do anything about it. It is like the virus. Neither will be going anywhere for a while.
11) Message boards : Number crunching : Uploading files fails (Message 62515)
Posted 26 May 2020 by Jim1348
Post:
I expect that the server uploads at some fixed rate. Whether that is a few people fast or a lot slowly probably won't affect the total time much.
They can tell us otherwise if they want to.
12) Message boards : Number crunching : Uploading files fails (Message 62502)
Posted 25 May 2020 by Jim1348
Post:
A shorter deadline, something on the order of 4 to 5 months, would seem to be in order as that seems to be how long the Researchers are willing to wait before they close the batches. There is no sense in running models that no one is ever going to look at because the batch was closed by the time you finished it.

Yes, that is my point too. The one-year deadline is just a waste of resources.
I think they do things just because that is the way they have always been done. For a science project, it is a bit curious.
13) Message boards : Number crunching : Uploading files fails (Message 62488)
Posted 24 May 2020 by Jim1348
Post:
I am uploading two zips now. They are a bit slow at 50 kbps, but at least going.
14) Message boards : Number crunching : New work Discussion (Message 62482)
Posted 23 May 2020 by Jim1348
Post:
Two machines is hardly statistically significant but I have one desktop and one laptop, however I see no difference in the error rates between the two.

I look at the errors on the other machines for work units that I complete successfully.
The difference is striking.
15) Message boards : Number crunching : New work Discussion (Message 62477)
Posted 22 May 2020 by Jim1348
Post:
OK, if it works for them, it works for me.
I see that the current batch of wah2 sold out quickly; I got two.
16) Message boards : Number crunching : New work Discussion (Message 62473)
Posted 22 May 2020 by Jim1348
Post:
As far as storing work for later goes, I don't bother with it. I have always believed that a fast turn around is more important to science than a slow, steady trickle of results no matter what projects you are crunching for. No matter what is being modeled faster returns will mean a faster evolution of the models. Getting better models designed means getting real world applications faster.

Yes, I would think so too. I sometimes wonder why the scientists have not insisted that the deadline be shortened. This is not a time or place for tradition. We are up against the wall on time.
17) Message boards : Number crunching : New work Discussion (Message 62469)
Posted 22 May 2020 by Jim1348
Post:
I STRONGLY DISAGREE! I have been running CPDN (almost exclusively) for more than 10 years, since the days of the BBC experiment. I have over 20,000,000 credits. All of this work has been run exclusively on laptops.

The time limit is the point, and also the error rate. If you can get past those, you can do them however you want insofar as I am concerned.
But on Windows, most of the errors I see are for too many suspends. I don't think that is from the desktops.
18) Message boards : Number crunching : New work Discussion (Message 62463)
Posted 22 May 2020 by Jim1348
Post:
So, where exactly are these ' huge queued stacks of WU's waiting on long due dates" sitting and sitting they are somewhere. In the old day's we used to squirrel away WU's on floppies or alternative media. Is this still going on?

I see a lot of suspends in the results files. It leads to a lot of errors, and lengthens the return time even for the ones that survive.
I assume that they are being done on laptops, which is the wrong place to do them I believe.

Better a reasonable time limit of 60 days (or 30 is OK with me).
19) Message boards : Number crunching : "No tasks sent" (Message 62417)
Posted 9 May 2020 by Jim1348
Post:
While we are on the subject, this is supposed to be an active hurricane season.
I hope that OpenIFS can make a contribution to the methodology of prediction.
20) Message boards : Number crunching : "No tasks sent" (Message 62415)
Posted 9 May 2020 by Jim1348
Post:
I am also curious about an inability to get any tasks. I have a new-ish computer purchased in 2019, running Windows 10.

OK, I can spare the moderators the trouble.
There are no Windows tasks (wah2) at the moment, and none on the horizon that anyone knows about.
https://www.cpdn.org/server_status.php

The only ones are for Linux now (32-bit by the way).


Next 20

©2020 climateprediction.net