1) Message boards : Number crunching : For the betterment of BOINC (Message 31653)
Posted 27 Jul 2017 by Profile Ageless
Post:
Hi BOINC Community,

Two weeks ago Richard Haselgrove and I joined a BOINC Work Group committee which researches how BOINC can be made more user friendly, easier for anyone to set up their own short- or long term project, and for the community to join in on those endeavours. The goal is to get more people to coding all parts that make BOINC (client, manager, web site, forums, projects, etc.), to test everything, to get them to set up their own projects, to make BOINC a future-proof and reliable brand that isn't dependent on any one person in particular.

We do know this is a big order and it won't be solved in a couple of weeks. So we'll work in the background together with some key people from projects/code developers to get things started. Eventually we will need answers from you as well, probably on a lot of things. But we'll start slow with a couple of small questions:

1. If there is anything at all you can change in/withdraw from/add to BOINC, what would that be? While we don't exactly look for enhancements or bug squashing, you may just let out anything you think would put BOINC on the map. For example to add social media inside BOINC Manager, or have certain add-ons integrated into the client (I am making these up, they aren't on the list (yet)).

2. Would you like to contribute to making BOINC better, or program for it, or walk the source code, or do anything to help the project forward? What has held you back thus far?

3. We'd like to get into contact with people who programmed for BOINC, but no longer do. Can you PM me or Richard on this, or contact me via email? Especially if you're one of the people in the Volunteer Developers section here.

4. We also like to get into contact with people who now voluntarily program for BOINC. Can you tell us why you decided to work on BOINC, how difficult it was to get into and what we can do to increase your involvement?

With thanks for any answers you have,

Richard Haselgrove
Jord van der Elst
2) Message boards : Number crunching : Faulty Computers or Modified BOINC ?? Huge Credits (Message 23751)
Posted 30 Nov 2011 by Profile Ageless
Post:
Your idea of flagging results that have a CPU time 10 times higher than the run time will not stop cheating because the cheaters will simply cheat both numbers and make the run time slightly higher than the CPU time, as it normally is. Your suggestion would not catch that cheat.

It would if the project updated the CreditNew version to the latest, which caps both the p_fpops value and the runtime value.

Going at the users with the ferocity you're doing isn't going to help. Witch hunt here, witch hunt there, alienating people who try to help by coming up with alternative ideas, shooting them down from the hip.

Perhaps it's time for you to take some rest, away from the daily crap dealt to you with BOINC. Try playing The Elder Scrolls V: Skyrim, it's addictive and before you know it, days have passed, days that you don't need to be snippy at people you don't know, never met and by going at them like this, never will meet.
3) Message boards : Cafe LHC : GridRepublic (Message 23696)
Posted 16 Nov 2011 by Profile Ageless
Post:
I just tried, it worked without much of a problem. Are you sure you used the correct password? What's the exact error message you get at GridRepublic? Have you tried asking at the GR forums?
4) Message boards : LHC@home Science : Why are w3mass tasks so fast? (Message 23670)
Posted 7 Nov 2011 by Profile Ageless
Post:
I am on vacation bit working when I can.

Bad boy! Back to the beach, you!

:P
5) Message boards : Number crunching : Resource share (Message 23669)
Posted 7 Nov 2011 by Profile Ageless
Post:
Also, do know that since Einstein has now a resource share of zero, that any work that it gets in will be done immediately, possibly even in high priority if there's a lot of it to be done, when you run with a large cache.

So it can even hypothetically happen that you have a lot of work for Einstein, that LHC has work in the mean time but that BOINC decides that you still have work enough in cache --or because work is running in high priority, that the downloading of more work is stopped until you're out of high priority-- and so that it doesn't download any work from LHC.

If you're totally unlucky it can even happen that you have so much Einstein work that LHC's work has gone before your BOINC allows more work to be downloaded again.

So just don't run with a large cache. :-)
6) Message boards : Number crunching : no more work? (Message 23619)
Posted 1 Nov 2011 by Profile Ageless
Post:
Not sure why you linked to my computers ?

He didn't. http://lhcathomeclassic.cern.ch/sixtrack/hosts_user.php is the general link to "your" computers. When I click on it, I see mine. ;)

7) Message boards : Number crunching : Long delays in jobs (Message 23596)
Posted 29 Oct 2011 by Profile Ageless
Post:
I've noted this to Igor. At this point my suspission is some server component is out of date and the settings noted in the boinc docs are for a newer version. We've determind now that the server software is over 1 year old. Be patient, this will all get sorted out eventually.

/home/boincadm/update_latest.sh

But that will probably break 10 other things. ;-)
8) Message boards : Number crunching : Long delays in jobs (Message 23530)
Posted 16 Oct 2011 by Profile Ageless
Post:
If the admins want to do that they can send me a zipped dump of the database and I'll write a script to comb the database for task crashing computers and computers that haven't contacted the scheduler for a while and send the owners an email.

And you can be trusted, how? Just on your pretty face? ;-)
Seriously, the admins shouldn't send anyone a dump of the database. It makes the project not very trustworthy if they do.
9) Message boards : Number crunching : Requesting new tasks for NVIDIA GPU? (Message 23509)
Posted 14 Oct 2011 by Profile Ageless
Post:
But why is Eric's system requesting GPU work? LHC Six Track does not have that option available in the Project preferences. I run GPU Tasks (NVIDIA) for other Projects but never see a GPU request for LHC Six Track.

Yours is probably already backed off to do that only once every 24 hours.
BOINC will by default ask every project that the user has attached to for Nvidia and ATI work, until the back-off is 24 hours.

It does this so the user doesn't have to change anything when the project decides to do work with a GPU. When they then release their application, it'll be picked up completely automatic when the back-off time is reached and depending of course on what GPU the user actually has. Otherwise it'll continue to back-off for another 24 hours.

To get to the 24 hour back-off period, there will have been several automated contacts. Do know that Updating the project manually will reset the back-off timer, and work requests for the GPU will be earlier again, increasing in time until it's back at 24 hours.

On projects where there is a GPU application, the back-end software will auto-detect this and enable the project preferences for the ATI and Nvidia GPU, depending on what application the project has released. Because this project has no GPU application at this time, the "Use ATI GPU" and "Use Nvidia GPU" options are disabled.
10) Message boards : Number crunching : Credit awarded calculation (Message 23489)
Posted 13 Oct 2011 by Profile Ageless
Post:
Decide for yourself as to who will waste their expensive energy charges and expensive machines for this?

Since you have no problem running an old Intel(R) Celeron(R) CPU 1.70GHz, a Pentium(r) II Processor, a couple of Pentium 4s all with HT, and an i5 with a ATI Radeon HD 5x00 series (Redwood) (1024MB) GPU, one would expect that you don't care about your electricity bill.

For if you did, you would:
a) shut down all those old machines that burn electrons more than they do work.
b) not run any work on a GPU, as that'll increase your electricity bill at least ten fold.
c) not run work on that many M (mobile) processors. Don't you know that laptops and notebooks eat more electrons per capita when doing BOINC than desktops do? Plus their cooling sucks.

Now then, things that make you go oooh. Your benchmarks on that i5 are way too low, even when it is a mobile. The machine was probably doing something CPU intensive when you allowed BOINC to make the benchmarks, or there's something else going on.

Your numbers in August:
Measured floating point speed 2281.02 million ops/sec
Measured integer speed 7232.77 million ops/sec

Now, for comparison, I have an i3, which has been doing these numbers quite consequently. My numbers should be lower than yours. My machine is not overclocked or anything. My numbers:
Measured floating point speed 2626.4 million ops/sec
Measured integer speed 9187.57 million ops/sec

That you claim too little for your perception is a cause of your low benchmarks.
11) Message boards : Number crunching : Tasks v530.09 crashing (Message 23456)
Posted 11 Oct 2011 by Profile Ageless
Post:
Well, either way, if I'm validating against a Windows computer with 530.10, I'm at a great disadvantage.

How'd you know? You have aborted all 530.10 work on all your computers without even testing one...
12) Message boards : Number crunching : Tasks v530.09 crashing (Message 23454)
Posted 11 Oct 2011 by Profile Ageless
Post:
However the current ifort version is actually rather slow until I get round to optimisimg it.

I'd like to disagree. :-)

Two of my last 530.09s:
CPU time 23271.08 seconds.
CPU time 27747.53 seconds.

This versus two 530.10s that have finished:
CPU time 11122.58 seconds.
CPU time 10511.39 seconds.

If you manage to decrease my run time by half to a third without optimizations, then please don't optimize the applications any further. ;-)

(I fully understand that not all tasks are the same length in run time around here.)

In the past we could test changes "in house" but not at the moment.

Well, you have us. You can send that work either to hosts you trust, or have a small group of us do some alpha work with feedback.
13) Message boards : Number crunching : Tasks v530.09 crashing (Message 23444)
Posted 10 Oct 2011 by Profile Ageless
Post:
530.09 <rsc_fpops_est> is <rsc_fpops_est>30000000000000.000000</rsc_fpops_est>
530.10 <rsc_fpops_est> is <rsc_fpops_est>120000000000000.000000</rsc_fpops_est>

Is there a reason why you increased the estimated run time by 4 times the original value? These tasks run for approximately 8 hours on my i3-530, but they're estimated to run a whole lot more, thereby making these task runs in panic mode (high priority).

Just because you went back to a lower form of instruction set does not mean everything will run that whole lot slower. ;-)
(With all those tricks, the LHC Classic DCF on my system is now completely haywire. Where the other projects that run have their DCF around 1.0, LHC has it at 2.9; I may consider resetting the project, so my DCF will reset to 1.0).
14) Message boards : Number crunching : Problem after updating computing profile (Message 23424)
Posted 9 Oct 2011 by Profile Ageless
Post:
Or you can set it through BOINC Manager->Advanced view->Tools->Computing preferences->CPU usage, as these preferences override any of the same preferences set by the web sites. That way it doesn't matter what any next new project makes of this preference, it'll be disabled on the computer using the local preferences.

Local preferences only work on the computer they were set on.
15) Message boards : Number crunching : Too low credits granted in LHC (Message 23421)
Posted 9 Oct 2011 by Profile Ageless
Post:
Is this aBOINC problem or an LHC@home issue ?

Neither. BOINCStats and all the other stats sites will have to import the statistics of the projects. I don't know of any project that constantly, every minute, updates the statistics in the format that the stats sites can read. I also don't know of any statistics site that will download real-time statistics off any project website.

So, the project will make new stats files once every 6, 12 to 24 hours, as it adds some overhead on the servers. The stats sites will pick up this information once every 6, 12 or 24 hours and use that data in their site.

But it won't ever be that credits you had just minutes ago will immediately show up on any of the stats sites.
16) Message boards : Number crunching : Tasks v530.09 crashing (Message 23400)
Posted 7 Oct 2011 by Profile Ageless
Post:
The Pentium 4 uses MMX, SSE and SSE2, as long as the operating system knows about these instruction sets as well. Windows 2000 does not support SSE2 or any instruction set thereafter. It only supports up to SSE.

I gave the instruction sets of SSE and SSE2 as examples, I hadn't checked all of Igor's posts to see what they were actually using. But at least that explains things further.

To be able to use SSE3, one needs a Pentium 4 Prescott CPU or better, or an Athlon 64 or better and Windows XP or better. SSE3 is also known as PNI (Prescott New Instructions).
17) Message boards : Number crunching : Tasks v530.09 crashing (Message 23397)
Posted 7 Oct 2011 by Profile Ageless
Post:
You always increment version numbers, so re-releasing 530.8 as 530.91 would be the next logical choice.

Ugh, it was 530.08 and 530.09, so the next logical choice is 530.10 ;-)
18) Message boards : Number crunching : Tasks v530.09 crashing (Message 23395)
Posted 7 Oct 2011 by Profile Ageless
Post:
Surely that's not too old and slow.

Both the Pentium III and the Athlon XP 2400+ can do SSE, but not SSE2 or above. So, if the 530.9 science application was compiled to use the SSE2 instruction set, whereas the 530.8 version was compiled to only use the SSE instruction set, then that's what causing these errors.
19) Message boards : Number crunching : Tasks v530.09 crashing (Message 23393)
Posted 7 Oct 2011 by Profile Ageless
Post:
Apparently, we should have kept the version 530.8 for older processors.
It is still possible, I have not removed them.

What would be the architecture designation for distinguishing the old and the new?

You always increment version numbers, so re-releasing 530.8 as 530.91 would be the next logical choice.

If you want to designate these to specific CPUs only, you'll need HR type 1, or to set up application plan classes.

Edit: But you can also ask yourself, is it worth it? Does this project have that many old CPUs attached? You can check that in the database. Or do you just not want to set a minimum CPU/minimum OS/minimum BOINC version as requirement? All questions you may answer for yourself. :-)
20) Message boards : Number crunching : Tasks v530.09 crashing (Message 23385)
Posted 6 Oct 2011 by Profile Ageless
Post:
Hold on, "Exit code 168" and "Exit code -168" are two different beasts.
BOINC errors are negative errors. Positive errors are science application errors.

The error here is "exit code 168 (0xa8)", thus a science app error.
The whole error is:
<stderr_txt>
forrtl: severe (168): Program Exception - illegal instruction
Image PC Routine Line Source
sixtrack_530.9_wi 00406C83 Unknown Unknown Unknown
sixtrack_530.9_wi 0040101F Unknown Unknown Unknown
sixtrack_530.9_wi 00657D33 Unknown Unknown Unknown
sixtrack_530.9_wi 006344EA Unknown Unknown Unknown
KERNEL32.dll 7C5989D5 Unknown Unknown Unknown

</stderr_txt>

I then looked up the "forrtl: severe (168)" part of the error and came to this thread on the Intel boards, where it says: "Turns out that the problem was caused by an older-generation processor not understanding newer instructions. The application had been compiled with the "Generate most optimized code" (/fast) setting, which implies /arch:host."

Probably something similar has happened here.
The "Genuine Intel x86 Family 6 Model 8 Stepping 3 746MHz", is a what? A Pentium III, an P2/3 model Celeron or a Pentium II Xeon? The 530.09 science application is probably compiled with an instruction set that these old CPUs do not understand. Hence the errors.


Next 20


©2024 CERN