Message boards :
Number crunching :
Far Far Too Much Redundancy!
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · Next
Author | Message |
---|---|
Betting Slip Send message Joined: 25 Jul 00 Posts: 89 Credit: 716,008 RAC: 0 |
> Just my opinion but this whole tread and argument has become redundant. > > Is there a way to just block an entire thread? > > Actually, there is, Use your freedom of choice and don't click on the title. A time honoured way of personal censorship. |
Scribe Send message Joined: 4 Nov 00 Posts: 137 Credit: 35,235 RAC: 0 |
OK, you persuade the Jockey Club to prevent anyone holding horse races with more that 3 horses, as there are only 3 places to compete for, the other horses are redundant, and I will start a petition to have no moe than 3 WU in a race! |
Betting Slip Send message Joined: 25 Jul 00 Posts: 89 Credit: 716,008 RAC: 0 |
> OK, you persuade the Jockey Club to prevent anyone holding horse races with > more that 3 horses, as there are only 3 places to compete for, the other > horses are redundant, and I will start a petition to have no moe than 3 WU in > a race! > LOL |
MJKelleher Send message Joined: 1 Jul 99 Posts: 2048 Credit: 1,575,401 RAC: 0 |
> > > I was wondering when the "if you don't like it you can leave" > argument > > would > > > surface. > > > > > And I would have hated to have you disappointed! Now, would you like to > > address the main part of the message: showing how other projects > are > > being deprived? > > > > > It certainly has not been shown to be false! > > > > Actually, it has. Regardless of SETI's validation policiy, Einstein gets > 40% > > of my CPU time. CPDN gets 10%, Protein gets 10%. The numbers say that > your > > assertion is in error. > > > > So show us how it is true. You seem to be the only one in this thread > with > > that information. Enlighten us. > > > > Resource share is irrelevant. By having a WU done 4 times instead of only 3 > 25% of whatever resource share is wasted. > Your assertion was that SETI's 4x redundancy was depriving other projects . This policy affects the use of SETI's allocation, which you (and no one else) deems to be waste. I call it their prerogative, in their protocol, to balance downloads and worked units against lost units that would have to be resent. How does it affect the other projects? |
Betting Slip Send message Joined: 25 Jul 00 Posts: 89 Credit: 716,008 RAC: 0 |
> > > > I was wondering when the "if you don't like it you can leave" > > argument > > > would > > > > surface. > > > > > > > And I would have hated to have you disappointed! Now, would you like > to > > > address the main part of the message: showing how other > projects > > are > > > being deprived? > > > > > > > It certainly has not been shown to be false! > > > > > > Actually, it has. Regardless of SETI's validation policiy, Einstein > gets > > 40% > > > of my CPU time. CPDN gets 10%, Protein gets 10%. The numbers say > that > > your > > > assertion is in error. > > > > > > So show us how it is true. You seem to be the only one in this > thread > > with > > > that information. Enlighten us. > > > > > > > Resource share is irrelevant. By having a WU done 4 times instead of only > 3 > > 25% of whatever resource share is wasted. > > > Your assertion was that SETI's 4x redundancy was depriving other > projects . This policy affects the use of SETI's allocation, which you (and no > one else) deems to be waste. I call it their prerogative, in their protocol, > to balance downloads and worked units against lost units that would have to be > resent. How does it affect the other projects? > If you can't see how any unnecessary redundancy in one project or more affects every DC project then go back to SCHOOL My original post requested that they get rid of unnecessary redundancy or if you like BUSY work. |
MJKelleher Send message Joined: 1 Jul 99 Posts: 2048 Credit: 1,575,401 RAC: 0 |
> > Your assertion was that SETI's 4x redundancy was depriving other > > projects . This policy affects the use of SETI's allocation, which you > (and no > > one else) deems to be waste. I call it their prerogative, in their > protocol, > > to balance downloads and worked units against lost units that would have > to be > > resent. How does it affect the other projects? > > > > If you can't see how any unnecessary redundancy in one project or more affects > every DC project then go back to SCHOOL > > My original post requested that they get rid of unnecessary redundancy or if > you like BUSY work. > I see. You have nothing besides your opinion that this level of redundancy is unnecessary. The rest is fluff. If you come up with any reasons, facts, I'm sure the development team will be happy to hear them. |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
> If you can't see how any unnecessary redundancy in one project or more affects > every DC project then go back to SCHOOL > > My original post requested that they get rid of unnecessary redundancy or if > you like BUSY work. ... and I'm a bit disappointed with the Ad Hominem attacks -- when you can't argue the merits, about all you can do is attack the person arguing against you. When you aren't attacking people, your arguments assumes facts that are not in evidence:
<li>Sending the work unit out four times is not necessary to get three results</li> <li>CPU cycles are a precious and limited resource</li>
|
Betting Slip Send message Joined: 25 Jul 00 Posts: 89 Credit: 716,008 RAC: 0 |
> > If you can't see how any unnecessary redundancy in one project or more > affects > > every DC project then go back to SCHOOL > > > > My original post requested that they get rid of unnecessary redundancy or > if > > you like BUSY work. > > ... and I'm a bit disappointed with the Ad Hominem attacks -- when you can't > argue the merits, about all you can do is attack the person arguing against > you. > > When you aren't attacking people, your arguments assumes facts that are not in > evidence: > >
> <li>Sending the work unit out four times is not necessary to get three > results</li> > <li>CPU cycles are a precious and limited resource</li>
|
MJKelleher Send message Joined: 1 Jul 99 Posts: 2048 Credit: 1,575,401 RAC: 0 |
|
Ertugrul Gokcen Send message Joined: 3 Apr 99 Posts: 39 Credit: 20,227 RAC: 0 |
> > If you can't see how any unnecessary redundancy in one project or more affects > every DC project then go back to SCHOOL > Well, I seriously think that it is YOU who should go back to school! I don't like the "Majority is always right" attitude, but if everybody in the thread except for you is saying something and you say the opposite, I think it is you who should sit back and rethink. Let's say you have allocated 25% to SETI. If they send extra WUs, which you claim steal time from other projects, but really DON'T, those WUs, which you claim steal time from other projects, but really DON'T, are crunched in that 25% slice of CPU time, and that DOESN'T affect the other projects which crunch in their remining 75%. You see, basic math, 75 + 25 = 100! It's a waste for SETI, though, but they don't seem to care, because they have more crunching power than they need, you may look at this thread. When they have more work or when other projects steal more and more CPU time from SETI, maybe the amount of work to be crunched will level with the amount of available CPU power, then they (project owners) may consider reducing the number of WUs per chorum. |
karthwyne Send message Joined: 24 May 99 Posts: 218 Credit: 5,750,702 RAC: 0 |
> You found a couple of people there to agree with you (and others who did not). > And still have not posted any answer to my question - how are other projects > being deprived? > @MJ he hasn't responded to anyone's questions, all he is capable of doing is attacking anyone who does not agree. i, like most ppl here, prefer Seti. if it were not for BOINC, i would give NO crunching to anyone but seti (maybe one PC running CPDN only), and i didn't until BOINC. SETI developed BOINC, so SETI is giving countless other DC projects a chance to get a bit of time on my computer. Seti created BOINC to get rid of their redundancy. how selfish of Seti (and if you need to ask, yes, that IS sarcasm) plus the fact as others have stated, redundancy at any project 'hurts' only that project. my resource share gives x% of time to my other projects. so, maybe the SCHOOLs i went to have better math programs, for Racing's math skills are severely lacking. Micah S@h Berkeley's Staff Friends Club |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
> he hasn't responded to anyone's questions, all he is capable of doing is > attacking anyone who does not agree. i, like most ppl here, prefer Seti. if it > were not for BOINC, i would give NO crunching to anyone but seti (maybe one PC > running CPDN only), and i didn't until BOINC. SETI developed BOINC, so SETI is > giving countless other DC projects a chance to get a bit of time on my > computer. Seti created BOINC to get rid of their redundancy. > how selfish of Seti (and if you need to ask, yes, that IS sarcasm) We do have to keep in mind that BOINC and SETI are different projects. SETI Classic was the first large-scale public distributing project, at least as far as I can tell, and I wish I could find some of the white papers where the project describes how successful Classic has been -- beyond their wildest dreams. BOINC was developed so that projects like SETI could share the tremendous resource (or perhaps feed the ravenous beast) that SETI Classic created. But, according to boincstats.com, SETI has 214,000 hosts crunching, and CPDN is #2 at 67,000. It's probably safe to assume that most hosts crunch one project, but it'd be interesting if one of the STATS sites could do a survey of the CPIDs and figure out how much overlap there might be.... |
Digger Send message Joined: 4 Dec 99 Posts: 614 Credit: 21,053 RAC: 0 |
I have to agree with Ned here on an important point: Racing, i don't believe you can make a blanket statement like "The vast majority of crunchers crunch for more than one project." Maybe that's what we see when we look around at all the stats pics, but the folks who only crunch for SETI aren't necessarily going to have those, and not everyone posts on the boards anyway. Many people just want to crunch for SETI, and always will. I think especially when we see Classic shut down for good, we're going to continue to see a huge influx of crunchers who just want to continue their SETI work and no other. Dig |
RandyC Send message Joined: 20 Oct 99 Posts: 714 Credit: 1,704,345 RAC: 0 |
> > The way that BOINC is designed, having 'excessive' redundancy in any > > particular project only deprives THAT PROJECT of computer time. Any > other > > attached project still receives the SAME AMOUNT of CPU cycles from each > user > > regardless of the level of redundancy the first project decides to > employ. > > . > > > > If your computer is crunching the 4th instance of a seti unit it can't be > crunching anything else so it has an impact on whatever projects you are > crunching. > This would be true IF and ONLY IF: 1. The forth instance of the Seti unit is not required for the quorum. 2. You are attached to another project (which some users have stated in other threads that they absolutely will not do...I disagree with them, but that's their choice) 3. Seti (or to make it generalized...the project sending excessive redundant WUs) would have completely run out of valid science work if they weren't sending redundant WUs. Highly unlikely with Seti, and probably wouldn't happen with Einstein either. |
MJKelleher Send message Joined: 1 Jul 99 Posts: 2048 Credit: 1,575,401 RAC: 0 |
> @MJ > he hasn't responded to anyone's questions, all he is capable of doing is > attacking anyone who does not agree. i, like most ppl here, prefer Seti. if it > were not for BOINC, i would give NO crunching to anyone but seti (maybe one PC > running CPDN only), and i didn't until BOINC. SETI developed BOINC, so SETI is > giving countless other DC projects a chance to get a bit of time on my > computer. Seti created BOINC to get rid of their redundancy. > how selfish of Seti (and if you need to ask, yes, that IS sarcasm) @Micah Yeah, I know, but it's a slow weekend, and I can get stubborn when somebody dodges my questions. 8-) I ran SETI exclusively until I switched to BOINC a couple of months ago. Einstein also interests me. CPDN and Protein are backups, Protein because one of my computers doesn't have the capacity to run CPDN at a small share and get it complete in time. Old box that's my file server for backups now. > so, maybe the SCHOOLs i went to have better math programs, for Racing's math > skills are severely lacking. Add logic skills, and I think you've got it pegged. > Micah |
karthwyne Send message Joined: 24 May 99 Posts: 218 Credit: 5,750,702 RAC: 0 |
> We do have to keep in mind that BOINC and SETI are different projects. > > > But, according to boincstats.com, SETI has 214,000 hosts crunching, and CPDN > is #2 at 67,000. > > It's probably safe to assume that most hosts crunch one project, but it'd be > interesting if one of the STATS sites could do a survey of the CPIDs and > figure out how much overlap there might be.... > @ Ned yes, of course. BOINC and seti are different. maybe i am incorrect in my understanding, but i thought it was the seti devs that thought up and developed BOINC? i also would love to see the percentages of hosts with one project. and using RAC, what percentages each project gets with the overall and one the shared hosts. (i know Zain is just loving us now ;) ) @MJ i am with you. there seems to be a lot of posts these days that just make me cringe. Micah S@h Berkeley's Staff Friends Club |
Astro Send message Joined: 16 Apr 02 Posts: 8026 Credit: 600,015 RAC: 0 |
BrainSmasher is right. When an individual or a group makes a choice for one thing over another, they generally choose the item that favors them. Seti made the choice to have 4 people crunch each WU because it reduces the time that a WU sits on the hard drive. Here's the part where he's right: It is a waste of the CPU cycles of one of the 4 crunchers. That being said, this is Berkeleys project. They will make choices that benefit them. If the costs were equal then I'd hope that they'd choose the option that favors us. It's their Project tony |
Astro Send message Joined: 16 Apr 02 Posts: 8026 Credit: 600,015 RAC: 0 |
> BrainSmasher is right. > OOPS.....Just realized I'm a dumb A**. Remove the name "BrainSmashR" and insert "Race Post Free Tips". Sorry, came out of the case of dumb A** after the 60 min edit limit. Otherwise, keep my comments the same. tony |
mikey Send message Joined: 17 Dec 99 Posts: 4215 Credit: 3,474,603 RAC: 0 |
> Race Post Free Tips is right. > > When an individual or a group makes a choice for one thing over another, they > generally choose the item that favors them. Seti made the choice to have 4 > people crunch each WU because it reduces the time that a WU sits on the hard > drive. Here's the part where he's right: It is a waste of the CPU cycles of > one of the 4 crunchers. > > That being said, this is Berkeleys project. They will make choices that > benefit them. If the costs were equal then I'd hope that they'd choose the > option that favors us. > > It's their Project > > tony > You are correct ONLY if the first 3 users return the unit in time and in-line with the other 2 users. If ANY one of the first 3 users is over the time limit, out of whack with the credit request, has a problem downloading OR uploading, etc. then the decision to send it to 4 users was/is correct. note...name changed to reflect what the OP meant to say. |
MattDavis Send message Joined: 11 Nov 99 Posts: 919 Credit: 934,161 RAC: 0 |
> > Where in my posts did I say "You're bad because you're from _____" that? > You call the project wasteful, and liken it to America, like we're some big homogenous entity. Vicariously you're calling every American on this board wasteful with a negative implication. Leave my country out of this, douchebag. I don't think you're anti-American. I think you're pro-asshole. ----- |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.