Welcome to MilkyWay@home

Posts by Link

1) Message boards : Number crunching : Long crunch time on new N-Body simulations? (Message 68743)
Posted 17 May 2019 by Link
Post:
Well, this isn't very surprising since the new application is using only one single core.
2) Message boards : Number crunching : So no way to select project campaigns anymore on the new server code (Message 68660)
Posted 3 May 2019 by Link
Post:
They are working on that.
3) Message boards : News : New runs of MilkyWay Nbody out (Message 68659)
Posted 3 May 2019 by Link
Post:
Eric Mendelsohn wrote:
How long ago were the preferences to opt out of Nbody or Separation last available?

They were available until the day when the server software was upgraded.


marmot wrote:
To deal with guaranteed N Body coming at a machine that had opted out, won't we now have to use an app_config with:
(...)
to reduce core count to 1 (or 2)?

No, the new n-Body application is single core anyway.
4) Message boards : News : New runs of MilkyWay Nbody out (Message 68634)
Posted 1 May 2019 by Link
Post:
Do you remember what the option for opting out of N-body units was called? Was it "Run test applications?"? I'm still looking into the website code to resolve this.
-Eric

No, it was something like:
Run only the selected applications:   Milkyway@home Separation: yes/no
                                      Milkyway@home N-Body Simulation: yes/no

If no work for selected applications is available, accept work from other applications?: yes/no
5) Message boards : News : New runs of MilkyWay Nbody out (Message 68631)
Posted 30 Apr 2019 by Link
Post:
On the other hand, what happens when you select both CPU and GPU? (That is, do you get only the GPU Separations then?)

Than you get Separation for GPU and Separation and n-Body mix for the CPU.
6) Message boards : News : New runs of MilkyWay Nbody out (Message 68627)
Posted 30 Apr 2019 by Link
Post:
How does a person opt OUT of the N-body work?

Just de-select "Use CPU". It is the only CPU project they have now.

No, "Milkyway@home Separation" has CPU apps as well, see applications page.
7) Message boards : News : 30 Workunit Limit Per Request - Fix Implemented (Message 68479)
Posted 5 Apr 2019 by Link
Post:
@ all having issue to get WUs here because of that 10 minutes thing...

Have you tried to set your cache to:
"Store at least 0.01 days of work" and whatever else additional and then set "network activity based on preferences" and not "network activity always"?

This should actually limit scheduler requests to once in 14.4 minutes... I think. I don't have a Milkyway compatible GPU to try that.

Just a thought until it's fixed on the server...
8) Message boards : News : New Server Update (Message 68331)
Posted 21 Mar 2019 by Link
Post:
This release will not include a MT version as we have decided it does not make sense for our project anymore.

OK, thank you, no MT version might be even better when mixing n-Body with other projects.
9) Message boards : News : New Server Update (Message 68321)
Posted 21 Mar 2019 by Link
Post:
In the Milkyway@home preferences the possibility to choose between milkyway and n-body WUs is missing. Maybe when you are fixing this, you could also implement the possibility to choose between mt and non-mt n-body WUs, that we have talked about here?
10) Message boards : Number crunching : What is supposed to be the total run time on N-Body Simulation tasks? (Message 68169)
Posted 18 Feb 2019 by Link
Post:
It has also stayed at 75.130% complete for a week or more now.

Than it is probably stuck. Restart your BOINC client (or the computer if that's easier) and it should continue. If the WU is over the deadline, check if the server is still waiting for your result, it the WU has been completed and validated for two other computers, you can abort it.

EDIT: is it this WU: 1717972366?
If yes, you can probably abort it (or try to be faster than the other person and get the credit if you want). In the feature, if a WU does not show any progress for few seconds, always check in the task manager if it's using CPU, if it's not, restart your BOINC client (or pause and resume the WU if you don't have the setting to keep them in memory when paused).
11) Message boards : Number crunching : more workloads. (Message 68147)
Posted 12 Feb 2019 by Link
Post:
There are server side limits on how many WUs you can have on your computer: Micromanaging CPU vs GPU Workunit Limits.
Also it's not a good idea to have 20 days of work stored on your computer, when the deadlines here at Milkyway are just 12 days. If you could get that many WUs, your BOINC client would abort many of them, since it wouldn't be able to start them before the deadline.
12) Message boards : Number crunching : What is supposed to be the total run time on N-Body Simulation tasks? (Message 68094)
Posted 2 Feb 2019 by Link
Post:
The runtimes for n-Body WUs are very different for each WU. I've seen up to 530000 CPU seconds on my i3 running at 1,4GHz. No need to abort them.
13) Message boards : News : Micromanaging CPU vs GPU Workunit Limits (Message 68067)
Posted 26 Jan 2019 by Link
Post:
By definition here at MW N-Body wu's are multi-core wu's.

No, there's also a single core N-Body application, see applications page.
14) Message boards : News : Micromanaging CPU vs GPU Workunit Limits (Message 68065)
Posted 25 Jan 2019 by Link
Post:
cpu and gpu is too much wats...need to choose something one.

You can choose that already in your MilkyWay@Home preferences.
15) Message boards : News : Micromanaging CPU vs GPU Workunit Limits (Message 68064)
Posted 25 Jan 2019 by Link
Post:
You do know you can turn each kind off individually in the settings...right?

Run only the selected applications MilkyWay@Home: yes
MilkyWay@Home N-Body Simulation: no


But that doesn't select between single-core and multi-core N-Body, does it?


Yes it does, if you unselect the "N-Body Simulation" units you won't get any more multi-core wu's, on the other hand if you unselect the "MilkyWay@Home" units then you won't get any more single core wu's.

No, it does not. With "MilkyWay@Home: yes", you will get single core MilkyWay@Home WUs (they are all single core), but not single core MilkyWay@Home N-Body Simulation WUs.
16) Message boards : News : Micromanaging CPU vs GPU Workunit Limits (Message 68052)
Posted 21 Jan 2019 by Link
Post:
Boinc itself won't fill the remaining cores with wu's because the setting is 3 cores per wu! That's a very inefficient way to run a pc!!

As descibed above the current mix of single core and multi core WUs is very inefficient as well, in some cases you may end up with just one core in use.


If you do I think alot of people will leave MW, I personally hate running multi-core wu's as they never seem to work well with other projects when you guys for example run out of work.

Well, the current situation is even more a reason to search for some other work for the CPU, I mean right now milkyway nbody does not even work well with itself. But my initial idea was to give the users the possibility to choose the application since I know the complains from number crunching forum about the mt application.
17) Message boards : News : Micromanaging CPU vs GPU Workunit Limits (Message 68048)
Posted 19 Jan 2019 by Link
Post:
Would that work?

Yes, at least with BOINC v7 clients (not sure starting from which version exactly), older clients will let the application use all cores. At least that what I have seen with v 6.10.18 and v6.12.34. The BOINC client I use now, v7.12.1, starts the mt application with as many thraeds as it is allowed to use, it will only use too many cores, if some single core tasks are running.

I'm not sure if the server will send the multi core application to computers that are allowed to use just one core, but probably yes since it was still sending single core WUs to my laptop even if according to my app_config they were using more cores than BOINC was allowed to use (this information is passed to the server in sched_request_milkyway.cs.rpi.edu_milkyway.xml). And the BOINC client also didn't care that the WU used 8 CPUs even if I allowed to use only 2.

So yes, I think that will work not be worse than now... I mean, if some problems occur, you can always return to two different applications.
18) Message boards : News : Micromanaging CPU vs GPU Workunit Limits (Message 68045)
Posted 18 Jan 2019 by Link
Post:
Can you explain a little more about what you want? Is there a reason you would prefer to run say 5 or 6 single-core runs instead of 1 multi-core run on 6 cores?

No, I'd like to run just one multicore nbody WU on all 4 cores of my i3 and no single core WUs at all.

With BOINC v6, after starting a single core WU the client wouldn't start another single core WU unless it's the next one in the queue and let the other cores idle, since there were not enough cores for the multi-core WU, which wants all of them.

BOINC v7 starts the multi core WU together with the single core, but with 4 threads (or for example 2 threads if I allow it only to use 50% of CPUs), so it's using actually more cores than it's allowed to, it runs the multi core taks like there was no single core task running. Not a real issue in my case, however I can imagine this can be an issue for people, who need to keep some CPU cores free to feed their GPUs.

If you need more details, just ask. I tried getting only multicore WUs with the help of app_config.xml, but it didn't work.

EDIT: like Jim1348 said, nothing wrong with any of the WU types, just mixing them isn't good.
19) Message boards : Number crunching : less cpu's per task (Message 68041)
Posted 18 Jan 2019 by Link
Post:
You can use app_config.xml, see Project-level configuration.

This should work:

<app_config>
<app_version>
<app_name>milkyway_nbody</app_name>
<plan_class>mt</plan_class>
<avg_ncpus>4.000000</avg_ncpus>
<cmdline>--nthreads 4</cmdline>
</app_version>
</app_config>
20) Message boards : News : Micromanaging CPU vs GPU Workunit Limits (Message 68040)
Posted 18 Jan 2019 by Link
Post:
Regarding n-body WUs: is there any hope, that we might get the possibility to choose wether we want to get multi core or single core tasks? Right now we have to use annonymous platform for this, but that means no automatic updates for the application.


Next 20

©2019 Astroinformatics Group