Message boards :
Number crunching :
Time has come 2 this
Message board moderation
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · Next
Author | Message |
---|---|
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
If you think the problem is staff, or a lack of funding, or bad servers, or not enough uptime, you're missing the point. BOINC starts with the idea that there is lots of research that needs to be done where there is, and likely never will be, anywhere near enough money. BOINC projects are by definition underfunded. It starts with the assumption that BOINC projects will use hand-me-down or inexpensive hardware. Instead of buying computing, it will use "waste" CPU cycles that would normally go to the system idle task. It assumes that the network is unreliable. When SETI@Home has the inevitable problems that result from the kind of funding available for "egghead" research that doesn't lead to new products (and make the researchers rich beyond the dreams of Avarice) we complain. We don't marvel at how well everything works given the resources. We never stop to think that we wouldn't be here at all if money was plentiful. Nope, we hold them to the same standards as Amazon.com or Google, but without the cash. If you disagree, don't take my word for it. Read the white papers on BOINC.Berkeley.EDU. |
-BeNt- Send message Joined: 17 Oct 99 Posts: 1234 Credit: 10,116,112 RAC: 0 |
If you think the problem is staff, or a lack of funding, or bad servers, or not enough uptime, you're missing the point. Thanks for restating what has already been said. Maybe they need it beat into their heads? Traveling through space at ~67,000mph! |
platium Send message Joined: 5 Jul 10 Posts: 212 Credit: 262,426 RAC: 0 |
It is more than likely that comunitions with S.E.T.I. happens before contact is made. If make contact it could be so important in exchange of techology and also to find out if they are peacefull.this project may take many years, so the down time is small and not that important, motor cars started off with more down time than they were running but they almost got it right now you finish most journey's |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
I can not completely get behind the "If it is up it is up, if not oh well" concept. If it is up, great! If it stays up, the guys at the lab can hopefully focus on moving the science forward. If not, well it is scrambling time for everyone. Staff, solicited and unsolicited helpful ideas.. which are also time consuming, review of what is needed to get things up, extra time (and therefore expense) in trying to return things to a working situation, fund drives, donations financial and hardware.. All in all a large game of whack-a-mole. Speed it up and set it to "Yakkety Sax"(Benny Hill theme). We are out here to help. The lab is skeleton staffed with Highly skilled scientists and programmers that have learned some about system administration, and by need some have gotten good at it. Not great, but good. Why not great? Well primarily because they have too much else to know/learn, do not have the hours, and have way too much on their plate. The hardware is no where near state of the art, absolutely not fault tolerant, hobbled together, and in some cases nearing the end of its life expectancy, or already beyond its life expectancy. The newer equipment should last a while, once tuned in should be more trouble free and hopefully allow them to put some extra effort towards moving the project forward, where we can move through the candidates(there are tons) and try to find ET. In the mean time, as it seems to be highly unlikely that new servers are in their budget without special attempts by US, If we can pull in a new 2 servers per year by shaking a tin can at people, we might let them be able to actually continue this search until it reaches its conclusion, whatever that conclusion might be. A perfect project might have unlimited funds, Bullet proof servers, Ample staffing, and oh yeah, they would not need us. Yay imperfections!! Janice |
-BeNt- Send message Joined: 17 Oct 99 Posts: 1234 Credit: 10,116,112 RAC: 0 |
I can not completely get behind the "If it is up it is up, if not oh well" concept. I don't either, but you really don't have a choice. You can donate as much as you want and get new hardware pushed to them all you want, but until they start getting donations or funding to staff the server closet with a admin 24/7 365, then you're kind of stuck. You can post about it as much as you want, scream until you are blue in the face, write angry attacking post, etc. but there is nothing that is going to change it. So if the server are down oh well, step back take a breather wait for them to fix it. No point in pulling out the demons just because it isn't going the way you expect it or want it to. So I take the "oh well" stance until they or I have the means to make it 'right'. One thing that keep popping into my head though, is what if they had the money for a super computer to crunch all this data. What would the RAC be on something like that? Man I want one! (BTW this is in no way an attack on you just general summarizer of the attitudes that have been displayed over the last month or so.) Traveling through space at ~67,000mph! |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
The servers have to wait until they can get to them. Hopefully we can help them need to get to them a bit less often. This worf failure was really poorly timed. Janice |
-BeNt- Send message Joined: 17 Oct 99 Posts: 1234 Credit: 10,116,112 RAC: 0 |
The servers have to wait until they can get to them. Hopefully we can help them need to get to them a bit less often. This worf failure was really poorly timed. Yeah Worf deciding to fail on use was a huge punch to back of our heads. If it would have stayed working for just a week or so longer they would have been in a much better position. But like anything else in life, there is never such a thing as not having a weak link. Traveling through space at ~67,000mph! |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
Not exactly. Most of what has been said is "it's too bad they can't get more funding" or "they're doing the best with what they have" or even "we'd be better off if more talented people ran things." What I'm trying to say is that SETI@Home has been working 100% within the basic BOINC design criteria, and that they have successfully met all of their commitments. Yes, successfully. Supercomputer-level work is being done without renting time on supercomputers. BOINC is trying to find a way to do great things on literally no budget. If you pay attention to the constant sniping at the project, then the whole BOINC way of doing things (and the whole idea of volunteer computing) is a failure. |
Josef W. Segur Send message Joined: 30 Oct 99 Posts: 4504 Credit: 1,414,761 RAC: 0 |
... I agree except for the last sentence. Social dissent does not constitute failure, and BOINC provides the basic forum infrastructure plus teams, etc. so that participants can interact if they want. BOINC is robust even when users withdraw compute capability more than new or returning participants and Moore's Law effects can counteract, and shrinkage is not failure. Joe |
-BeNt- Send message Joined: 17 Oct 99 Posts: 1234 Credit: 10,116,112 RAC: 0 |
Sorry these threads are getting long and I'm starting to lose track of what's in each thread. I had posted those exact, or close, ideas in the other thread. Doh, long days and short time spans are killing me! Traveling through space at ~67,000mph! |
soft^spirit Send message Joined: 18 May 99 Posts: 6497 Credit: 34,134,168 RAC: 0 |
If it was not for SETI, there would be no BOINC. Boinc simply took what was developed within SETI and spun it out to other applications. Janice |
perryjay Send message Joined: 20 Aug 02 Posts: 3377 Credit: 20,676,751 RAC: 0 |
Actually, BOINC was an idea whose time had come. Without BOINC we would have a bunch of little projects begging for computer time. Trying their best to get the word out that they were there and needed help. SETI wasn't the first. I can't remember right off hand which project was but DA got this burr under his saddle and came up with the idea of a program to manage all the little projects so that people could go to one place and find what they were interested in. Since he was right their on the hill with the SETI crew, he used them to test his idea. We still are a test bed for BOINC and are probably the largest of the projects. PROUD MEMBER OF Team Starfire World BOINC |
1mp0£173 Send message Joined: 3 Apr 99 Posts: 8423 Credit: 356,897 RAC: 0 |
... The key word in the last sentence is "if".... I didn't say anyone should pay attention to all of the tantrums, the whining, and all of the people who seem to want to burn the staff in effigy. In fact, all of the stupidity that is commonly on exhibit is just a rather boring demonstration of how people can form opinions without any reference to facts, set performance goals well in excess of what the project promises, and then demand that the project meet criteria that are impossible without a couple more zeros in the budget. If anything, the way the new servers were funded is a stunning example of how extremely successful SETI@Home really is. Then there are the people who say "where is the science?" without thinking that the result is either "1" (found it) or "0" (not yet), and there is no way to say "we're 87.4% of the way to finding ET. So, Joe, I think we're in agreement. |
ScarabDrowner Send message Joined: 13 Sep 03 Posts: 90 Credit: 456,378 RAC: 0 |
...then demand that the project meet criteria that are impossible without a couple more zeros in the budget. There's an extremely easy way to add however many zeros you need to the budget: stick them out in front. :) |
tbret Send message Joined: 28 May 99 Posts: 3380 Credit: 296,162,071 RAC: 40 |
Okay - I noticed on the charity appeal letter that there wasn't a whole lot in it to motivate me to throw money. (I'm speaking generically.) So what would *you* say (that is "you the cruncher community") that Matt and Eric et al need more than anything else? Something that would be "life-changing" and change things permanently for the better. Something that maybe they'd never even ask for? What would that (whatever it is) cost in round numbers? Not money, but "a thing" like... a stipend for a CS grad student to come sit with the servers over the weekends for a year; two more new servers; rent on better digs for a year; the big data pipe; etc. |
kittyman Send message Joined: 9 Jul 00 Posts: 51478 Credit: 1,018,363,574 RAC: 1,004 |
Okay - I noticed on the charity appeal letter that there wasn't a whole lot in it to motivate me to throw money. (I'm speaking generically.) I think it's too soon to determine that. Things have not yet been to a stable state since the startup of Oscar and Carolyn to know where the next bottleneck lies. I do suspect it would be with that 'big data pipe' to the outside world, but right now Matt has indicated a limiting factor in the radar blanking of data before being able to split and send it out...... And getting a grad student to babysit the servers should not be necessary, pending further hardware failures. I think we may have to wait just a bit before we know what our next target would be. "Time is simply the mechanism that keeps everything from happening all at once." |
perryjay Send message Joined: 20 Aug 02 Posts: 3377 Credit: 20,676,751 RAC: 0 |
Okay, I'm going from memory here but as I recall, Matt had mentioned the gigabit line dedicated for SETI quite awhile back in the tech news forum. Hurricane Electric is already supplying 1 gig but it stops at the campus. To get the fiber line needed to supply it up the hill to the SETI guys would cost more than we just did in the last two fund drives. The number $38,000 sticks in my mind for some reason. I'm pretty sure that's the figure he mentioned. Seti can't just go to their provider and say run us a line, it has to be done through the campus IT department and there is no way they are going to let that much go out of their budget. The gigabit line just done was a campus upgrade already budgeted but was for the whole lab. SETI would have to come up with the cost of running a new line if they wanted their own gigabit line. PROUD MEMBER OF Team Starfire World BOINC |
KWSN-GMC-Peeper of the Castle Anthrax Send message Joined: 28 May 99 Posts: 274 Credit: 6,936,182 RAC: 0 |
I'm coming in late and I hope I've managed to correctly pick up the subject here. I feel the need to point out that people working in the Sciences have NEVER been known for their people skills...quite the opposite in fact. Over the years, the communications i've seen with the core staffs of all the SAH and BOINC projects do nothing but confirm this centurys old truth. Nobody was more pissed off than I was when they shitcanned all the work of us original participants at the beginning of BOINC. That's right, we enable their work and they ignore us and treat us like strangers much of the time. In return though, over the centuries, that's gotten us electric lights, hot running water and indoor toilets. I'm satisfied at the tradeoff. To add a thought...while the distributed computing aspect undoubtedly gets some respect in the community, the actual crunching of SETI data probably doesn't. Let's face it folks, we ain't gonna hear CRAP with the antenna at the bottom of the atmosphere...now if we could get one on the farside of the moon we might be talking some results...Anybody who believes THIS data is going to find ET probably also believes in the intrinsic good of humans, the Easter Bunny and Trickledown Economics. If you don't touch it, you can't break it. ; |
Matt Lebofsky Send message Joined: 1 Mar 99 Posts: 1444 Credit: 957,058 RAC: 0 |
This sort of discussion/debate comes up from time to time on these boards, and with good reason because the sentiments on both sides (or however many sides) are valid. Personally, I'm always at complete odds with how little public relations this project has, which is why I'm driven to inform as much as I can. To clear up a couple things (I scanned some of these posts, sorry if I missed yours): 1. I actually live 6 miles away from the lab via surface streets or 9 miles away by highway. Basically a drive without traffic one way is about 20 minutes. My record time for coming to the lab in an emergency and returning home victorious is about an hour. Usually it's like two hours as some babysitting is required after kicking whatever needed to be kicked. As a fun aside, it turns out by some small miracle the lab (which is located not on the campus proper but in the Berkeley hills) is actually in my line of sight from my kitchen window (about 4 miles away as the crow flies). I dream someday of installing a zip line which would make the commute from the lab to my backyard MUCH faster. 2. I'm not officially a scientist, though I act like one from time to time. I'm really a software person, my nerd life starting with many hours writing raw machine language on an Apple II. That said, given all the experience I have dealing with every aspect of internet-era IT I feel fully qualified to be a webmaster, or database administrator, or sys admin, or whatever IT position in any situation of any size. I live in the Bay Area and have lots of friends in the biz working on ostensibly prime-time stuff but I have yet to hear about any "pro" project that seemed beyond my scope of expertise. It's just computers. 3. The whole intern/volunteer thing seems worth entertaining at first glance, except I can't help but wonder who is going to train and manage these people. It sure as hell ain't gonna be me. I don't have the time. Even the smartest IT people out there would need a few months to ramp up on all the details about how our (quite non-traditional and complex) project works before they are useful. Yes, sometimes all we need is for somebody to come up and push a button and me, Eric, and Jeff are too busy. That's when we call Dan or Dave. 4. We didn't ditch any science from SETI@home classic. Yeah, a lot of people were angry about various aspects of the painful transition, and so we lost a number of old participants. But we still have the results they processed. 5. The gigabit link issue: that actually has been getting some positive progress. Turns out the rest of the lab is catching up with the times and suddenly is more willing to pay for a general gigabit link for everybody to use down the hill. We still have to pay for our own bits, and therefore will keep the HE link, but we can share this other link (as long as we're not maxing it out, which we won't be). Long story short, we still need some special hardware to share this link, but the total cost of the project is now far less than before. I don't have exact numbers (nor know what hardware it is we need offhand - I'm out of this loop). The time frame is still slow - central campus computing support has been helpful but their staff is vastly reduced and our project is low priority. 6. As always I'm overwhelmed by the generosity of many of y'all in wanting to get us the hardware we need. We actually have a new system being assembled/tested coming down the pike, and trying to get oscar and carolyn up to speed. Let's let the dust settle on these - then we'll have a better understanding what our needs are. Anyway... I dig the frustrations of many people. I'd be annoyed too if I were you. I do have three other active careers (in the computer gaming industry, the music industry, and the food service industry) so I know how things work in the world. As a professional, I know the excuse that we are a scrappy understaffed academic project isn't that great of an excuse, given the needs of our large volunteer base. But we work with what we got. - Matt -- BOINC/SETI@home network/web/science/development person -- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude |
perryjay Send message Joined: 20 Aug 02 Posts: 3377 Credit: 20,676,751 RAC: 0 |
Matt, As always, thanks for posting. It makes it a lot easier when we have some idea what's going on. It is really good news that the rest of the lab is getting interested in getting the Gigabit link. I'm sure if you post what your share of the cost and how much the new equipment will run you, we will find it. That's been a bottleneck for many a year. PROUD MEMBER OF Team Starfire World BOINC |
©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.