Welcome to MilkyWay@home

Posts by Astromancer.

1) Message boards : Number crunching : Lowering required cpus (Message 62903)
Posted 28 Dec 2014 by Astromancer.
Post:
Thanks Kwartet!

I'm not sure why they set the required cpus so high when the GPU app uses almost no CPU for this project.

This 270x I just got is disappointingly slow. I'm pretty sure my old 5870 crunched faster than the new card. But it's a gaming computer and it's better for gaming. I guess you win some and you lose some.

One other thing while I'm thinking of it. IIRC the updated apps hosted at Arkayn's site have been integrated into the project for some time so there is no need to get any op apps at this point?
2) Message boards : Number crunching : Lowering required cpus (Message 62899)
Posted 27 Dec 2014 by Astromancer.
Post:
Hey guys,

I know a lot of people say to leave a core open for GPU computing, but I have to say that I've found that to not really have much of an effect on crunching speed and I'd rather have that core running POGS and have MW run a bit slower on my GPU since POGS is CPU only.

Now the problem is with .94 required cpu cores by default if I try to run Einstein on my iGPU which required .5 cpu cores it stops a POGS unit.

I'm sure there is a way to write an app_info file however it's been a while since I've done any crunching and don't remember for the life of me how to properly set one up.

So if anyone could suggest a guide or has a file for just ATI GPU tasks that I could modify it would be appreciated!
3) Message boards : Number crunching : What is the cause of these 'validate errors' (Message 62898)
Posted 27 Dec 2014 by Astromancer.
Post:
I just got back to crunching recently and am running into the same problem as you guys are. However I'm only running one 270x and am getting the random tasks not validating.

I'm not sure if it was a certain batch of WUs or what, but I'm down to just one invalid currently. I'll keep an eye on it and post back if it goes back up.
4) Message boards : Number crunching : We have some new long running WU's!!! (Message 46115)
Posted 9 Feb 2011 by Astromancer.
Post:
The thing I'm wondering about is in the news thread about the new NV app Matt says something about WU's being created by accident that were longer...

We may have been too soon in our rejoicing...
5) Message boards : Number crunching : concurrent blocks / nvidia .. question :) (Message 46089)
Posted 9 Feb 2011 by Astromancer.
Post:
Besides the setting no longer doing anything, you are correct that you need a compute capability 1.3 or higher card to run the project. So you won't get any work for them from MW.
6) Message boards : Number crunching : We have some new long running WU's!!! (Message 46088)
Posted 9 Feb 2011 by Astromancer.
Post:
Just noticed this a few mins ago and my first thought was that the server was borked! lol

Very cool to have these.

Observations are, slightly higher GPU usage than previously when using the stock app with all CPU cores loaded (I could be wrong here but it's not lower on my 5870). Checking out the completed tasks I'm using overall less CPU time than running a few of the shorter WU's to equal a longer one. And I'm not seeing anything odd with memory usage, it's sitting at about 45mb for the program.

7) Message boards : Number crunching : Host with WAY too many tasks. (Message 45842)
Posted 28 Jan 2011 by Astromancer.
Post:
CTAP,

The thing you seem to keep missing is that the generation of new WU's is done by using the ones returned.

Server sends out 500 WU's, as it gets them back it's using those that are the best fit to generate new WU's that should give an even better fit. Lets say you can cache 200 WU's here and they take about a minute and a half to crunch. That's 5 hours that those WU's have missed out on getting better.

Now I'm sure your going to say something about CPU's and having the WU's for more than 5 hours. Yes, it's true that they do. But why take the GPU's that are advancing things faster and then slow down the entire process by people processing out of date data?

Who knows if giving us larger caches would in fact speed up the science by letting us crunch through the server downtimes or slow it down by most people crunching data that was not as relevant by the time they got to it. I'd imagine the latter is the case or they would have given us a larger cache or done away with the limits completely.
8) Message boards : News : Database Problem (Message 45841)
Posted 28 Jan 2011 by Astromancer.
Post:
Seems like you guys got everything back to where it should be. I'm seeing teams again and the stats sites have incremental updates.

And before Friday even so you can leave to party early! ;) Enjoy!
9) Message boards : Number crunching : New MH@Home user - checking in..... (Message 45759)
Posted 25 Jan 2011 by Astromancer.
Post:
To elaborate on what Beyond said.

Every WU is worth a fixed ammt of credit, much like at SETI. SETI iirc removed the "Claimed Credit" from where you look at tasks, MW could do the same since it doesn't mean anything here.
10) Message boards : Number crunching : New MH@Home user - checking in..... (Message 45756)
Posted 25 Jan 2011 by Astromancer.
Post:
Welcome!

For the most part there are no wingmen. There are times when a quorum is required just to make sure your computer isn't going crazy. But it's the exception not the rule.

The server should automatically send out SSE2 optimized apps if your CPU supports it, there are SSE versions as well. Between that and the apps being updated nobody has done anything else with them that I know of yet at least.

Same goes for the OpenCL nVidia app. So everything that auto downloads is the best there is currently for Linux (That I know of at least).


Though to ask a question, what distro / driver version are you using? I tried the nVidia app with Ubuntu / distro distributed drivers the other day with pretty poor results. I had a lot of errors that didn't show up until the WU was done.
11) Message boards : Number crunching : Host with WAY too many tasks. (Message 45751)
Posted 25 Jan 2011 by Astromancer.
Post:
Blurf,

I know of course the Matt / Travis will be the people who do anything with planning or the like. I just don't like being a person with only questions and all, though being someone with suggestions can be just as bad I suppose! Ha!

I'm glad that there is / will be a plan on how to find and what to do with those who are causing potential problems by hoarding the work.


I just wanted to know that the post asking not to do it wouldn't be the only thing being done. And I'm glad that there is a bite to back up the bark.

So thanks guys for all the work you do keeping this all running for our crazy hobby! And I hope everything sorts itself out nicely with this current crazyness.
12) Message boards : Number crunching : Host with WAY too many tasks. (Message 45747)
Posted 24 Jan 2011 by Astromancer.
Post:
Matt,

It's great that you guys posted that, but do you have a plan to check for / deal with people who are somehow bypassing the cache limit either in the form of being able to get more WU's than 6 per core or using the Blox method of creating many caches?

We the community can only do so much due to the low number of WU's that need to be validated as well as the fact that they are removed from the database rather quickly. And with the multiple cache method, having ones computers set to private would prevent a member of the community from seeing that it's been done.


So I'm not just asking questions but perhaps helping with a solution...

Perhaps use a script that checks to see if any computers have more than X WU's that are "In Progress" or more than X identical machines? Then you can deal with them as you feel necessary.

Or a per account limit on the maximum number of tasks in progress for all computers on an account? Perhaps something like 1000 WU's "In Progress", which would make someone need to have 166+ CPU cores to be able to cache more. I doubt many have more cores than that on their account.

A script that forced merged hosts at some random interval, say between 1 and 10 days or something so people couldn't plan for it, might be used to fix anyone using the multiple cache method as it would merge all the computers into one?

-TJ
13) Message boards : Number crunching : Host with WAY too many tasks. (Message 45643)
Posted 20 Jan 2011 by Astromancer.
Post:
In case you think it is, it's not my computer. Only reason I saw it was due to the fact that I was checking on who was crunching some of my older pendings.

Since the number of tasks they have is going up, while I'm still waiting for them to crunch something from the 13th... I'm wondering if it's some kind of error like the SETI Ghost WU's or something of the like.

But no matter if it's an error or someone cheating the system to get way more WU's than they should, it's not good.
14) Message boards : Number crunching : Host with WAY too many tasks. (Message 45638)
Posted 20 Jan 2011 by Astromancer.
Post:
So I was going through my WU's that were sitting around waiting for wingmen to report back and came across http://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=247482.

At first I thought perhaps it was a bug and the tasks would go down as they timed out / were reported. But it went from having 74xx last night to 76xx when I just checked... So obviously something is wrong here.

Not being sure what to do, I figured I'd post it here and someone else would know / a mod / project admin would see it or whatever.
15) Message boards : Number crunching : app_info - no n-body error... (Message 45281)
Posted 24 Dec 2010 by Astromancer.
Post:
Thanks Arkayn,

That's what I started to think when I saw I'd sometimes get tasks on a request. I myself am doing Collatz as that's my backup project.


@Werkstatt
I have the CPU tasks turned off, but it seems that if there is only N-Body work I get that error no matter what.
16) Message boards : Number crunching : app_info - no n-body error... (Message 45272)
Posted 24 Dec 2010 by Astromancer.
Post:
So I'm only running the GPU app and using an app_info to tweak how it runs.

Now since about 7:30 I've been getting:

Milkyway@home Message from server: No work sent
Milkyway@home Message from server: Your app_info.xml file doesn't have a version of MilkyWay@Home N-Body Simulation.

Is it that there is no work other than n-body work, or did the server just randomly decide not to like me?

Thanks in advance!
-TJ
17) Message boards : News : OpenCL for Nvidia available for testing (Message 44761)
Posted 7 Dec 2010 by Astromancer.
Post:
After I posted the last one I figured you would need more details since the deleter is deleting things right away. (Looks like I was right) So I went about getting them for you.

One other thing I noticed on my system is that the OpenCL tasks sit at 100% with the clock still going for about 30s.

I'll PM you with all the info so I don't make a huge post full of data useless to anyone but you and Travis.
18) Message boards : News : OpenCL for Nvidia available for testing (Message 44755)
Posted 6 Dec 2010 by Astromancer.
Post:
Matt,

I downloaded the WU's all today and within about an hour of each other. I did 4 OpenCL then 3 CUDA before posting and another few CUDA after posting (With the same type of run time seen).

The memory bandwith usage struck me as odd as well which is why I posted it. I was running GPU-z to watch what was going on a bit to make sure it wasn't say using 50% of the GPU core or something and noticed that the "Memory Controller Load" was reading at 50% or over. I've only ever seen that with SETI before.

I'll give a try with 48.1 and see if anything different happens. If there is some kind of test WU I can run through the command line or the like to help you out any, I'd be more than willing to do it.
19) Message boards : News : OpenCL for Nvidia available for testing (Message 44749)
Posted 6 Dec 2010 by Astromancer.
Post:
With 48.0 I found that the runtimes are over 2m longer than with the CUDA app on my GTX260. 15:40 - 16:00 for OpenCL and 13:22 - 13:31 for CUDA.

The memory bandwith used on the card is up to about 50% as well which is a HUGE jump from CUDA (0%). As well as having a larger system memory footprint. Just found that interesting, it doesn't particularly concern me.

And it also uses more CPU time than the CUDA app, though I didn't notice the cpu being used while I was watching the task manager. Does it use most / all of the cycles at the start or end of the WU or something?
20) Message boards : Number crunching : Status: "Validate Error" and still getting credit? (Message 44038)
Posted 21 Nov 2010 by Astromancer.
Post:
One of my team members checked his and apparently all of his have validated. *shrug* Who knows.


Next 20

©2021 Astroinformatics Group