Welcome to MilkyWay@home

Posts by Palo M.

1) Message boards : Number crunching : RSS feed problem (Message 36388)
Posted 10 Feb 2010 by Palo M.
There is a problem with RSS feed of project news (http://milkyway.cs.rpi.edu/milkyway/rss_main.php). It is caused by unescaped ampersand character in <title> tag. If you look at it in Opera, you will see error very clearly described. Looking at the RSS with Firefox displays no content at the moment (as the first entry in RSS is faulty) and aggregating this RSS feed in Drupal also leads to error...
Can this be corrected?
2) Message boards : Number crunching : WU Credits (Message 5130)
Posted 31 Aug 2008 by Palo M.
Yes, validation of open source code is a big problem. We would have no idea if the results returned to us were valid or not. So that's that's a factor for leaning toward closed source. We would like to have it so people can submit the code changes directly to us so we can validate there's no malicious activity going on. However, the BOINC framework doesn't currently support this. It's basically very decentralized and doesn't check with our server to see if you're running good binaries or not.

You can still work this out with open source:
You will publish only the computation code and simple test/validate code (so contributors can validate the updated code by themselves before sending you the proposals).
But the binary produced by you will contain additional code for signing of the result (the signing key will NOT be released to public). Validator's task will be to check the signature.
Then participants can contribute to your code by sending you updates, but if they attempt to use their private binary in the BOINC, signature will be missing and therefore not validated.

If you would mention such contributors on the project page, this kind of honour could be still motivating... maybe even more motivating than additional credits which the optimizer can get today.

But, signing of the result requires additional effort and it still does not guarantee the 100% good results (e.g. overclocked machine could return result, which is numerically incorrect, but signature will be valid). I still believe that quorum 2 is better way of validation...
3) Message boards : Number crunching : WU Credits (Message 5116)
Posted 30 Aug 2008 by Palo M.
Well, after having a break I visited the forum again and I'm really surprised where the discussion is going to...

Limiting the credit based on time? Hahaha! Hahaha! Hahaha! Desperate!

It took me few moments to get idea what Crunch3r (or anyone else) COULD do. Remember, Crunc3r is skilled coder. So, if the project is restricting the credit based on computation time, Crunch3r can modify the source to simply report full time. Of course, this itself would not be sufficient...
If his app is calculating let's say 25-times faster than stock app (I'm not investigating exact number, it can be any number), he will add code to sleep (at the end of calculation) for 25-times the calculation really took. Just don't do anything. Only then the application will finish - and report 25-times more than it really took, so it will seem to be stock application (based on time). But, what to do with so much time sleeping (= no CPU load)? Well, let's start VMware... 25 virtual machines in fact... and every virtual machine will do the same... so at the end, the load on physical machine will be 100%... And all together it will be reporting 25x of the results than single machine with stock app, using 25 virtual machines on 1 physical machine and getting full credit of 25 machines. And for everybody it would appear like farm of 25 computers and there will be no way to remotely distinguish these virtual machines from any other physical machines...
There are several other variation to this idea as well, since BOINC itself is also open-source.

Happy limiting, if it's worth for Crunch3r, he will do it to you and feast on credits!
Or, more probably, he'll just delete his optimized source, go another project and let you compute for 20 years what could be accomplished in 1 year ]:->
4) Message boards : Number crunching : I've had enough !!! (Message 4849)
Posted 20 Aug 2008 by Palo M.
Emanuel wrote:
I do not wonder. It's the "magic" of optimization of code to speed. Try some and you will see yourself. If you do not want to try, just think about very simple case: different speed of different sorting algorithms.

There is no magic in computer sciences. The greater the difference in speed between two applications getting the same result, the more radical the differences between those applications generally are. This might make it a lot more difficult to check, and Travis has a point about holding off on the optimizations until the new code is out. However, I do hope they can work together in the future and help us all contribute as much as possible to science.

I quoted the word magic. It shouldn't be taken literally.
And, I agree with you that optimized code could be quite different from original one.
But, there should be a way to check correctness of the code easily. NOT ONLY because of possible optimization, but mainly because of further evolution. After each update of your code you should re-run the tests to make sure that you accidentally did not spoil something.
I made some optimized applications for a few other BOINC projects (although mostly just re-compilation without any source code change, to produce faster binary for my processor)... The first thing I always did were the tests to make myself confident about correctness of the build. And I still closely tracked the first few results reported to project.
Of course, at the beginning of software development it's important to produce something which really runs. But then, after the first version, there should come further steps. Tests are important in software development generally, in BOINC even more. If there are no tests available to ensure correctness of application (which seems to be the case for MilkyWay at the moment), then I can say that the software is still in the pilot phase and there is still long way to go...
5) Message boards : Number crunching : I've had enough !!! (Message 4845)
Posted 20 Aug 2008 by Palo M.
Conan wrote:
I do wonder how a computer (no matter how overclocked), can take 4 minutes to do a WU that I take 5 1/2 hours to do, that is extreme optisation I must say.

I do not wonder. It's the "magic" of optimization of code to speed. Try some and you will see yourself. If you do not want to try, just think about very simple case: different speed of different sorting algorithms.
6) Message boards : Number crunching : I've had enough !!! (Message 4841)
Posted 20 Aug 2008 by Palo M.
Thank you for explaining the background, Travis.
Some my notes to whole topic (not addressed to specific person):

  • Although I'm technically oriented, in BOINC world I've learned that distributed computing is a lot about psychology. So Travis may have preference to do a science work for MilkyWay, but somebody from project should think about the cheating, as in the crowd there will always be somebody thinking about cheating. And if there will be real cheating, other persons will discover it and the flame war will start. And there could be false accusation or at least controversy... None of this is good for project (from psychologic point of view).
  • The optimization may be done for good motives. But optimizer may make unintentional mistake and the results will not be correct. Also, there could be some problem related to hardware - often there are unstable oveclocked machines used, which makes results incorrect. This also supports the opinion that quorum should be 2 instead of 1 and validator should eliminate incorrect results.
  • IMHO the whole credit system is wrong concept (or unfair, depending on point of view). Look at the definition of cobblestone - it's about amount of time spent on reference computer, without taking into consideration how the time was actually spent. If the employer would pay me for time spent in office without any other conditions, without asking whether I do something meaningful or useless, I would sleep in the office everyday.
  • Now the location of the source code is known. But now I do not want to do my own compilation, as at the moment it seems it would not help the project (see my 2nd remark above). I'll wait until project is ready for that... I hope the day will come soon.

7) Message boards : Number crunching : I've had enough !!! (Message 4789)
Posted 18 Aug 2008 by Palo M.
Crunch3r wrote:
obviously not hard enough as it's being "hidden" in plain sight ....

Wow! It is available, but I would say it is well-hidden under camouflage net... Now tell me the secret, how did you find it? Did you browse paper (the last one on project home page, but some other were also published in news section) and just sniffed in the parent directory?

BTW, the directory doesn't look like SCM. Is it really official stuff?
8) Message boards : Number crunching : I've had enough !!! (Message 4781)
Posted 18 Aug 2008 by Palo M.
zeitgeistmovie.com wrote:
As has been posted many times. The source is open and available to everyone.

Yes, it has been posted many times that source is open and available to everyone. But I have never actually seen any location of the source code. UncleVom was asking for the location of source code here as well, but there was no reply to him.
I was looking for source myself very hard, but it must be very well hidden, as I did not find it...

So is the source really open and available to everyone? Or is it just common false belief?

©2024 Astroinformatics Group