LHC@home

I've just got a wu in 1 month and a half.

The Boincwoman.
Member og Boinc@enmark

I'm not sure if I followed all that bowlingguy300 so I'll just "respond" to what I think I understand, if you know what I mean.

looking at the preferences for LHC, I noticed that the cache setting isn't in the LHC@home preferences but rather in the general preferences and apply to all attached boinc projects...

Sorry, I thought that was a known. I should have added that to one of my posts not so long ago.

...to me over caching doesn't make sense on this project....

Increasing the cache reduces the chances of getting LHC work units. It will be filled with other projects work units and is unlikely to take on more work. There is even a chance that BOINC Client may wait up to 10 days to check, by then any LHC work units will be gone. NOTE: bowlingguy300, I'm not highlighting to insult you but I would like others to notice that point.

I don't think I deserve any more units then anyone else, be nice to get some once in awhile....

I feel that way too. I have 5 PC's running and just a little while ago there were over 15,000 LHC work units waiting. One of my PC's was looking for work at that time and got about 10 work units, the other four got nothing so far and probably wont. They seem to be crunching a lot of work units for 3 or 4 other projects just at the moment. Don't know what to do about that, my cache is at 0.25 at the moment, LHC resources are at 20% to 40% depending on the PC. Seems to me that should have grabbed plenty but it didn't and there are still 14437 work units in progress.

It's hard for anyone to get work units here.

---

Click here to join the #1 Aussie Alliance on LHC.

Lets all watch this video Seed: Seed Short Film: Lords of the Ring An exclusive tour of the underground accelerator at CERN led by the scientists who work there. and see if we can find out where the wu's are disappearing to so quickly....

Unfortunately Quick Time stuffed me around again. It seems ever time I use it, it wants a new bit that it can't get! Not on the server this time :-(

Sounded good though. ;-)

---

Click here to join the #1 Aussie Alliance on LHC.

Yea, I saw this video and it's just great. Both scientificaly interesting and amusing.

For SC>Mike Mitchell: Try Mplayer instead for playing MOVS.

---

This guy got 50 new ones today!

http://lhcathome.cern.ch/show_host_detail.php?hostid=90486

Let's beat him up and take his work units!

---

-----

I got about 30 (between two machines) today, and they'vre chugging along. They seem to all be "short ones", and take a bit over an hour each.

With my resource share, basiclly, little but LHC will be done until they're gone, and that will likely be by midday tomorrow.

---

This guy got 50 new ones today!

http://lhcathome.cern.ch/show_host_detail.php?hostid=90486

At this time of posting, he's completed 14 of those. So he should be finished them by the end of the week at that rate. Dual CPU's must be nice.
Some work units in February, Some in June, Some in July.
With 50,000 or so user id numbers it's a random draw whether you get work units or not.
I wouldnt accuse this computer of 'hoarding' because it's missed a few work unit runs since February. In my opinion, it's just luck.

---

I'm not the LHC Alex. Just a number cruncher like everyone else here.

Of course it's luck. I thought me saying that we should beat him up and take his work units would have been a clue that I was joking -_-

---

-----

How this make does that it have there work only during a few minutes?
I finally could charged 2 works since 1 month and, 2 minutes afterwards, there is not more.
It is somewhat exaggerated!!!

---

In 00:34:43 and 00:43:40, it is already finished!
Let us hope for how the following do not delay too much! And that the situation come back like before.
It there does not have so a long time, work was certainly rare but DISTRIBUTED!

---

Perhaps this is a Social or Psychological experiment and not one of Physics...

I just see the paper on "Volunteer Responses to Contrived Shortages in an Altruisic Computing Environment..."

Does anyone have any insight into how CERN intended to build and configure the LHC **before** they joined BOINC?

I posted the response below to the Synergy board as there were complaints there too, mostly quanderies. Exellent people, but not as personally involved in particle physics as we are. My apologies if this goes "under" your heads :)

Why LHC Starts and Stops

I've seen several comments, complaints and scratching of the head as to why the Large Hadron Collider (LHC) at home puts up a bunch of workunits just to run out of them.

Weee'll it has to do with the very nature of the project. Projects like SETI keep running as long as they have the time on Arecibo to poke their noses into the universe. They gather rheems of data that we crunch. With few exceptions (server crashes etc.) they supply a constant streem of data.

LHC isn't doing this kind of research yet. What LHC is currently doing is designing and tuning their system. We supply numbers, LHC puts them in their simulator and says "Wallah, we can put that magnet together!"

If SETI were designing antennas, receivers etc. they would be doing the same type of work as LHC. Every time LHC gets in data from us, they have to evaluate it and then go to the next step. Once there, they reevaluate and send us more data to crunch. In essence this is a step by step process.

"Ok", you say, "I'll bite. Just what are they doing?"

We smash 2x4's through walls to see if walls survive tornados. We fire bullets into Kevlar vests to see if our soldiers can survive. LHC smashes atoms into other atoms to see...what will happen.

A linear accelerator is just like a gun. You fire a bullet/atom at a target/another atom. These particles are fired and accelerated by alternating electromagnets. The principle drawback is size. You can only get so much energy out of one magnet and the faster you go the bigger the magnet because in the firing cycle the magnet only gets used once.

This problem was fixed by making the accelerator round. Then the firing can go by the magnets lots and lots of times, acquiring more energy every time. The energies are astronomicaly higher than a linear accelerator.

Well here comes problem number one. Put a hole through a tennis ball and put a string through it. Then twirl it over your head. Travels in a circle don't it? Now let it go. Physics 101: An object in motion tends to stay in motion. The ball travels in a straight line when released. Your energy holding it (and the string) are the only thing that kept it going round. Minus gravity Mother Nature prefers things travel in straight lines.

So first we tune (read "Send WUs to us to crunch.") the magnets to pull and push electrons and other particles to horrendous speeds. But now we have to adjust those electromagnets to make that particle travel in a circle like a doughnut. You and I crunch again. LHC applies this and they are further in their design.

Just when you thought they had it figured out: They put a second doughnut right on top of the first one. This one has the electron/target going in the opposite direction. You got it! We crunch again. What LHC did to save money is put the two tracks next to each other and use one set of magnets for both. Now they have to figure out how to pull and push both going in opposite directions. We crunch. Uh oh, tricky part. What to do when they are passing each other in the same magnet? We crunch again.

I'll be brief. Frequency, voltage, amps and a number of other variables have to be taken into consideration so the magnets fire at just the right time (mili, nano seconds). Yup, start crunching again.

Remember when I said let the string go and the tennis ball will go straight? Well somewhere in here LHC lets the particle go and it smashes into the target. Same place every time. Think you can let go of the string and have the ball go the same direction every time? I hear you crunching again.

Now here is really the neat part. At 14 TeV (terra electron volts) an electron can be speeded to nearly the speed of light. What happens when two cars each going 40 mph collide head on? They hit at 80 mph. What happens when two particles traveling in opposite directions at nearly the speed of light hit each other? A collision at nearly twice the speed of light!

Last I looked, they haven't planned this one. Oops, they actually did it on low level in 1971.

http://public.web.cern.ch/Public/Content/Chapters/AboutCERN/WhatIsCERN/WhatIsCERN-en.html

"Lacy put down that disintegrator gun! Oh yeah, it's always funny until you put someones planet out!"

Andy

I'll admit, I like to crunch for numbers (as well as to further CERN). I consider LHC my pet project. I just think of biology projects as yucky. I always hated biology. I took LHC off of my three dinosaurs. It didn't take too long to see they couldn't take it. I did add three more P4 HTs to the two I was using only to not see enough WU's show up since. I am still waiting to see them strut their stuff.

I don't know what this "Cache" thing is. I just checked my SETI specificatons and didn't see Cache mentioned anywhere. Is it the percentage of memory used?

I will admit, I do stuff WU's but I can get them done in 2 days. I didn't know there was a problem until recently when I saw the low amount of WU's out there and can't grab any of them. Just now, there were about 500 and within minutes it went to 11,000 but all were taken. If I slow down CERN, I will surely back off, but I haven't seen any of my numbers busting the ones members here have set out.

Wasn't this whole thing about the computing power we bring to CERN? If we weren't here, just how much further "behind" would they be on only their processors.

One way or another we are a valuable asset to them, not a liability.

PS: Right before people went on vacations etc. for the summer, distributed computing put out over 400 Teraflops. The fastest supercomputer in the world puts @280 Teraflops, making us the fastest computer in the world.

---

I am a Geek. We surf in the dark web pages no others will enter.
We stand on the router and no one may pass.
We live for the Web. We die for the Web.

.... What happens when two particles traveling in opposite directions at nearly the speed of light hit each other? A collision at nearly twice the speed of light!
....
.

hmmm....

Sort of like setting your
" Connect to network about every..."
to 20 (twice the max allowed...)

....
....
I don't know what this "Cache" thing is....
....
.

It seems so!

Take a look at your general preferences for:
Connect to network about every
(determines size of work cache; maximum 10 days)

Bigger numbers allow you to get more work.

But, you should set your "Connect to network about every"
to no more than one-tenth.
That is the "fair" thing to do...

---

To Galeon 7

Thanks for posting that excellent 'Why does LHC stop and start' post. Very informative.

You mentioned two tracks in that post. If we are crunching 'six track' I suspect we will be reconfiguring and recrunching for a very long time?

D

---

Remember when I said let the string go and the tennis ball will go straight? Well somewhere in here LHC lets the particle go and it smashes into the target. Same place every time. Think you can let go of the string and have the ball go the same direction every time? I hear you crunching again.

Not so. The beams do indeed run parallel within the same tunnel. There is no 'exit point'. Rather, at points around the ring the beams cross, and it is at these points that the real experimental science takes place.

As the beams cross there is an occasional collison. I say 'occasional' for a reason. Most particles in the beam just sail straight through and continue round the accelerator. 'Most' is very nearly all - maybe all of them most of the time. But, because of the immense speeds generated any given particle will cross the collison points millions of times per second, the possibility of a collision within a sensible time frame becomes workable.

There is a separate modelling program (Geant4) that models the behaviour of particles and their products at the collision point. This was the subject of a BOINC porting exercise, but the computing environment required will defeat most home computers.

---

Gaspode the UnDressed
http://www.littlevale.co.uk

[quote

....
....
I don't know what this "Cache" thing is....
....

It seems so!

Take a look at your general preferences for:
Connect to network about every
(determines size of work cache; maximum 10 days)

Bigger numbers allow you to get more work.

But, you should set your "Connect to network about every"
to no more than one-tenth.
That is the "fair" thing to do...[/quote]

Then I would assume that .04 is ok?

---

I am a Geek. We surf in the dark web pages no others will enter.
We stand on the router and no one may pass.
We live for the Web. We die for the Web.

Mike, you know, teaching physics 101 to grad students really sucks :) Yeah, what I said does imply an exit point.

Crossing the beams (magnetic deflection) in Alice, Atlas, LHCB and CMS are the exits I am referring to. With a drawing it is easy to see, but without, a little hard to envision that something inline can actually be an exit point.

You know, they tried doing this in Ghost Busters and almost destroyed the space/time continuum don't you?

And it's all your fault!

Note to self. Never try to explain the difference between a quark and a qwerk to a physicist and a psychologist, because then they will get married and bring about the end of the universe. Philisophically speaking of course.

---

I am a Geek. We surf in the dark web pages no others will enter.
We stand on the router and no one may pass.
We live for the Web. We die for the Web.

TY Dora. Six-Track is only the name of the program they're running. It has no real connecton with the actual number of tracks in the LHC. In several places they are at minimum space so six tracks wouldn't fit equipment wise and the redundancy would be a waste.

The length of the old LEP and the LHC may be primarily the same at 27 Km, but the equipment in the LHC is significantly bigger.

It has to be when you're out to destroy the universe, like Mike. :)

---

I am a Geek. We surf in the dark web pages no others will enter.
We stand on the router and no one may pass.
We live for the Web. We die for the Web.

.... What happens when two particles traveling in opposite directions at nearly the speed of light hit each other? A collision at nearly twice the speed of light!
....
.

hmmm....

Sort of like setting your
" Connect to network about every..."
to 20 (twice the max allowed...)

....
....
I don't know what this "Cache" thing is....
....
.

It seems so!

Take a look at your general preferences for:
Connect to network about every
(determines size of work cache; maximum 10 days)

Bigger numbers allow you to get more work.

But, you should set your "Connect to network about every"
to no more than one-tenth.
That is the "fair" thing to do...

My machines that are always connect (usually) have a cache of 0.1 days. My machines that can only connect once a day have a cache setting of 1.0 days. If it were any less, they would not be able to crunch all of the time (they are attached to enough projects so that I don't really care which project they crunch for at any given time). The reason that I say usually 0.1 days is that I work on the CPU scheduler and work fetch algorithms - sometimes I need to test some boundary conditions (and I have not yet gotten an LHC result when the cache was set high on the dev machines).

---

BOINC WIKI

I think they are 'restricting' the workunits from the hog by doing smaller batch sizes. They run out of work in just an hour or two. Big Cache helps more now. It knocked two of my computers out because one has a internet connection that has to compete with Emule P2P (because I'm not getting rid of it) and the other that is only online when I swap the vzw pc-card to it. So only my work computer will ever get work now because it has a internet conncection that is good for it begging for work every few minutes.

---

My... LinkSite | Blog | Pictures

Quote fron Rosetta@Home RSS-Newsfeed:
"Jul 15, 2006
A note from David Baker: We desperately need as much CPU power as possible for the next two weeks--there are more than 25 CASP targets due, including some that are our best shots at really high resolution models. Frustratingly, we won't be able to do anywhere near as much sampling as we had planned for these proteins as there are so many coming due near the same time, and thus can't really expect the accuracy we had hoped for. So if it is at all possible for you to increase your rosetta@home cpu time for the next two weeks please do--it will make a huge difference for our collective efforts!"

Well, it looks like there are people desperately needing your crunchin'.. mighte be developing DC-apps to better investigate the way your (and your relatives and friends') proteins fold and make you (and your relatives and frineds) either healthy or (badly) ill is not as worthwhile and attarctive as developing the right tuning for some superconductive electromagnets driving hadrons around to inevstigate what the universe is, but a volunteer cruncher should be a generous person by definition.. so why not giving David a chance for next the 4 days nad let the LHC-hungries try to feed thir caches up with the few hadron-WU the CERN-people are releasing these days?

---