Message boards :
Number crunching :
Too low credits granted in LHC
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · Next
Author | Message |
---|---|
Send message Joined: 3 Dec 05 Posts: 11 Credit: 992,093 RAC: 25 |
University Flashback! Assembly was only part of the fun. The architecture of the 'processor' made things interesting also. Back in my days on submarines, I worked on CDC-1604 mainframes with RD-281 disk drives. The CDC-1604 was a second generation Cray, and the RD-281 was the original Winchester disk drive, the one where the head carriage (9 pounds) was moved by a hydraulic system powered by a Cessna aircraft 120 PSI hydraulic pump. The CDC-1604 didn't have a CPU. It didn't have a main processor as such. The little beastie had a 'subtractive adder' as it's central processor. The subtractive adder, by itself, was about twice the size of 2 current day tower PC's. Programming it was not a task for the faint of heart. There were no add, multiply, or divide instructions available. You had to add by subtracting, hence the name subtractive adder. Take a simple problem, like adding 2 plus 3. You start by getting the operands in the registers as usual. Then, negate the operand in the accumulator. Make sure you use the correct instruction, since one's compliment is a direct bit by bit inversion, but the two's compliment handled the sign bit as well as adjusting the value in the LSB. You now have negative 2. Then, subtract the other operand (3) giving you negative 5. Now negate the accumulator again, giving you positive 5. Multiply and divide operations were exponentially more complicated. After writing a few short programs in some of the training courses, it became very clear to me why programmers in the 'days of old' always had a 500 count bottle of aspirin on their desk and wore glasses thick enough to be bullet proof. They were also easily recognizable because they had either turned completely gray or gone bald sometime in their mid to late twenties. |
Send message Joined: 17 Sep 04 Posts: 40 Credit: 293,269 RAC: 0 |
And I remember when I started Assembler programming, on a PDPD-12, and was assigned a whopping 4,096 bytes of memory. No way I could build bulging muscles moving that much memory around. -ChinookFöhn |
Send message Joined: 18 Sep 04 Posts: 143 Credit: 27,645 RAC: 0 |
Here are some points from the documentation, I can't find "CreditNew" exactly as other people are refering to, but there is a secion called "The third New credit system", I assume this is it. Keith, surf to http://boinc.berkeley.edu/trac/wiki/CreditNew. :-) Jord BOINC FAQ Service |
Send message Joined: 2 Sep 04 Posts: 209 Credit: 1,482,496 RAC: 0 |
Here are some points from the documentation, I can't find "CreditNew" exactly as other people are refering to, but there is a secion called "The third New credit system", I assume this is it. Thanks, but that is where I got the info from, I didn't see that the only place it says "CreditNew" is in the url. Kinda lame if you ask me, which you didn't. |
Send message Joined: 18 Sep 04 Posts: 143 Credit: 27,645 RAC: 0 |
It's called "Credit system as of 5/2010" in BOINC software development. You can also check out CreditProposal, which has a lot more possibilities, which not necessarily will be added. Seeing how this project is running SSV22168, it has CreditNew (22168 is 9 August 2010). Whether or not it's activated, that's another thing. Jord BOINC FAQ Service |
Send message Joined: 2 Sep 04 Posts: 209 Credit: 1,482,496 RAC: 0 |
It's called "Credit system as of 5/2010" in BOINC software development. Igor tells me it is. They installed a new version, whatever was at the time of the downloading, and only imported only the database for users from the old project, so no code was copied from old project. Also to it is a different server and different o/s than what was at QMC/UK, hence some of the problems that have appeared as alomsot everything is new, except us old users. Learning curve too as everyone behind the scenes has to learn also how this all works. Igor also said the validator was rewritten using the sample validator that came with that latest download, so i'm pretty sure everything is fresh stuff, and not some (unknown) artifact that may have been on the project when it was hosted at QMC/UK or the original 7 year old project code. I kind of just assumed if they installed the latest, CreditNew was by default in use. Is there some specific server setting to actually active CreditNew, I can ask them to double check and see ? |
Send message Joined: 17 Feb 07 Posts: 86 Credit: 968,855 RAC: 0 |
Hello, Credit new or credit old, a lot of credit, to less credit, and what more... I see the following: A WU of Einstein@home (wich can be marked as a perfect project with active admins, and they are fokkowing the rules) takes approx. 22.000 seconds on a quad and gives 251 credits. LHC@home WU runs approx. 28.000 seconds (on the same quad) and gives around 110 credits. Algoritms, cpu cycles, old pc, new pc, credit junckie, or only crunching for science, I think this has to be equal, or even a bit more credit from LHC. Greetings from, TJ |
Send message Joined: 25 Jan 11 Posts: 179 Credit: 83,858 RAC: 0 |
There is no such thing as a perfect project. There are no rules that govern credit so how can Einstein be following the rules? It makes just as much sense to state that Einstein is giving too many credits and should decrease their credits to agree with Sixtrack's credits. |
Send message Joined: 3 Dec 05 Posts: 11 Credit: 992,093 RAC: 25 |
I found something interesting... go the the QMC website at: http://qah.uni-muenster.de/projectinfos.php And you'll find the following at the bottom of the page: Credits: That's a small part of the situation I was attempting to address with my question regarding credit grants. If you know how many credits you will grant for a particular WU, then cheating is very nearly eliminated. But, that brings us right around in a complete circle to the start of the conversation, - what is a fair standard for calculating credit? Hoo boy.... I never ask the easy questions, do I? :-) |
Send message Joined: 17 Feb 07 Posts: 86 Credit: 968,855 RAC: 0 |
We can loose the credits, see how many crunchers stay around for science. Greetings from, TJ |
Send message Joined: 3 Dec 05 Posts: 11 Credit: 992,093 RAC: 25 |
We can loose the credits, see how many crunchers stay around for science. Deja vu... Now, what about those who think the faster computer should get more points for completing the work faster? Or, what about those who think that the slower computer should get more points because it spent more time crunching? I'd be willing to bet that if all the BOINC projects stopped granting points altogether, we would find out relatively quickly who was crunching for the warm fuzzy feeling of contributing something worthwhile to science, and who was crunching solely to get the most points. It does make you wonder. With cheating being the issue that it has become, why do people find it necessary to cheat at all. It sometimes looks like the cheaters have the attitude that 'whatever you do, I can do better'. If the credit system were abolished, would these people have a nervous breakdown, pack up their toys and leave, all the while screaming about how unfair it is that they can't be the best anymore? |
Send message Joined: 17 Feb 07 Posts: 86 Credit: 968,855 RAC: 0 |
It does make you wonder. With cheating being the issue that it has become, why do people find it necessary to cheat at all. It sometimes looks like the cheaters have the attitude that 'whatever you do, I can do better'. If the credit system were abolished, would these people have a nervous breakdown, pack up their toys and leave, all the while screaming about how unfair it is that they can't be the best anymore? You could be very right J Henry Rowehl. At MW there was a big issue in spring with a lot of outages. They promised 48 hours of double credit. We got it in summer, but complaines that men didn't cruch because the heath. Even the 48 hours couldn't be made dus a new issue. They promissed the make that up. An now people are asking once in a while where the double credit days stay. An for the ones who stick at the project wel it is fair, but you see the credit juckies come. We will see what happens. Greetings from, TJ |
Send message Joined: 18 Sep 04 Posts: 143 Credit: 27,645 RAC: 0 |
Is there some specific server setting to actually active CreditNew, I can ask them to double check and see ? I couldn't find one in config.xml (the server's project configuration file). But when I see talk to one of the developers, I'll ask. Jord BOINC FAQ Service |
Send message Joined: 2 Sep 04 Posts: 209 Credit: 1,482,496 RAC: 0 |
Is there some specific server setting to actually active CreditNew, I can ask them to double check and see ? I got cut off before, was at work and had to cut my post short. I tried to find the PM but I must have deleted it. I remember Igor telling me though that they used the validator example with the new download they did for the project back here at the LHC, that it was rewriiten using that sample and thet is calls some function, like credit_dev (? not sure on that name) so to me it sounded like they use some default boinc mechanism and not any formula they dreamed up here. --- Anyway whatever formula is used, there will always be some that are happy with it and some that are not. We can change it again and different people will speak up as being unhappy. Do away with and and the same, another group of unhappy people. No matter what is tired/used there never will be 100% that accept it. |
Send message Joined: 18 Sep 04 Posts: 143 Credit: 27,645 RAC: 0 |
I got an answer back, those devs never sleep it seems. ;-) CreditNew is always on, by default. It can be disabled, but that requires some hacking of the server code. There is also an alternative credit option, but this is for projects that use a nonstandard notion of job, used by e.g. CERN's Test4Theory that runs through a virtual machine. Jord BOINC FAQ Service |
Send message Joined: 25 Jan 11 Posts: 179 Credit: 83,858 RAC: 0 |
Anyway whatever formula is used, there will always be some that are happy with it and some that are not. We can change it again and different people will speak up as being unhappy. Do away with and and the same, another group of unhappy people. No matter what is tired/used there never will be 100% that accept it. That's true but that doesn't solve the problem of how to grant credits and how many to grant. There has to be a decision on which group(s) we're going to satisfy and which group(s) we're not going to satisfy. I think the crunchers who forget that the primary purpose of BOINC is to facilitate research and not to handout credits should be ignored. And I think projects who cater to those credit whores should be excluded from the community or even disabled somehow. Milkway is the first project I would disable if it were up to me. If they don't want to play nice then they shouldn't be allowed to play at all. Double credit days is a blatant rejection of the long standing principle that hosts should receive credit for work done. Double credit days gave credit for work that was not done. I know for a fact that Dave Anderson talked Heidi (Milkyway admin) out of double credit days at first but it's obvious from T.J's report that she went back on her word and did it anyway. That project should be booted immediately. I think Sixtrack should push this issue and make it a BIG concern for EVERY project. I think Sixtrack should award 1,000,000 credits for every successfully completed task regardless of elapsed time or CPU speed. For every unsuccessful task they should award 500,000 credits because at least the host tried and that's worth something too. Let's see how Milkyway likes taking a dose of their own medicine. |
Send message Joined: 17 Feb 07 Posts: 86 Credit: 968,855 RAC: 0 |
That's true but that doesn't solve the problem of how to grant credits and how many to grant. There has to be a decision on which group(s) we're going to satisfy and which group(s) we're not going to satisfy. I think the crunchers who forget that the primary purpose of BOINC is to facilitate research and not to handout credits should be ignored. And I think projects who cater to those credit whores should be excluded from the community or even disabled somehow. Milkway is the first project I would disable if it were up to me. If they don't want to play nice then they shouldn't be allowed to play at all. Double credit days is a blatant rejection of the long standing principle that hosts should receive credit for work done. Double credit days gave credit for work that was not done. I know for a fact that Dave Anderson talked Heidi (Milkyway admin) out of double credit days at first but it's obvious from T.J's report that she went back on her word and did it anyway. That project should be booted [b]immediately No, no, no that is not it. You have to read the whole story. In spring we run for almost 48 hours with validated results, but to teh action of the admins, al credit was lost. So all crunchers at that time did not get any credit at all, but they sent in good and validated and thus usefull results. That will be made up. And when that started it went wrong once more. What prof. Heidi Newberg (professor in physics no administrator) wanted was something else and is and will never be done. Greetings from, TJ |
Send message Joined: 4 May 07 Posts: 250 Credit: 826,541 RAC: 0 |
No More Credits Granted for LHC@Home Geneva - The LHC team at CERN has just announced that they will no longer grant credits for work units on the LHC@Home project. Credits will be granted to a new Debit Card and if you want to see how much of a balance you have in your Debit Card you will have to pay a $5.00 per month per CPU fee. This is in line with the new banking policy in the United States (and probably coming to a bank near you) of collecting fees for Debit Card usage. End of sarcasm Frankly, I crunch for the Projects I'm involved in because what they are doing interests me. As far as I'm concerned they can just stop granting credits. I recall some years ago there was an ongoing battle about credits on the SETI project. I haven't the slightest idea as to how much effort it took the Project's people to make everybody happy but I'm sure they could have been doing things a lot more productive. |
Send message Joined: 24 Nov 06 Posts: 76 Credit: 7,919,502 RAC: 3,223 |
That project should be booted immediately. Booted from what, exactly? There is no central authority. Each project is completely independent, and can do whatever they choose. Dublin, California Team: SETI.USA |
Send message Joined: 25 Jan 11 Posts: 179 Credit: 83,858 RAC: 0 |
Booted from the BOINC community. Stats sites could possibly be convinced to not import their stats, just for a start. There IS a central authority named Dave Anderson and he can do whatever he wants too, within the limits of the law, of course. Dave is under no obligation to let anyone use BOINC. He can make use of BOINC contingent upon any condition he wants including the condition that they issue credits in a decent, responsible fashion. He isn't obligated to distribute it as open source. He can distribute it as closed source and build into the executable a function that contacts his server and sends an encrypted request for permission to continue. If he/software elected to not respond with a 'yes' then the project would be effectively suspended. Reasonable projects will agree, in a EULA, to that condition and the consequences of not complying. Existing projects that don't agree to the EULA won't get vital upgrades. Existing projects that agree to the EULA will export encrypted stats. Stats sites will be given a binary (closed source) that decrypts those stats and requests permission to do so. If Dave refuses permission they can't decrypt/import stats. Legal implications? I don't think any of the stats sites has money to contest such a scheme. If a project is funded/hosted by a university or research institute then the project has to ask them for permission and funds to pursue litigation. That'll fly like a manhole cover when the governing authority at said institution learns that all the project has to do to not be suspended is give out reasonable credits. I'm describing a possible scenario here. I am not saying this is going to happen or even that I have heard it might happen. I believe that the above scenario does not agree with Dave A's current philosophy but since some projects don't use BOINC the way Dave intends I can easily see Dave changing his philosophy and punting rogue projects off the playing field. It can't happen soon enough, IMHO, so I will be shopping my proposal around to all the projects and to Dave. |
©2024 CERN