Just reading an interview with Jos Engelen, Chief Scientific Officer at CERN, about the LHC.

Each {LHC} experiment will produce 10 petabytes of data a year and will need 100,000 CPUs to analyse it all"

According to Jos Engelen, CERN are colaborating with partners like the European Union and Leading IT companies to create the worldwide LHC Computing Grid project - LCG

So what happened to BOINC then? LCG certainly doesn't look like it's a one off.

---

Interesting, they can't keep enough work on hand to feed the number of users they have now.

Interesting, they can't keep enough work on hand to feed the number of users they have now.

They're talking about the data generated when it's actually up and running smashing tiny bits of sub-atomic particles back into last tuesday.
That and looking at the colo(u)r of quarks, mu-mesons or whatever.

---

I had heard that they were impressed with the power of Boinc, especially when compared to all the Rental time on Supercomputers they've been paying for. I guess each project knows best what their needs are.

I had heard that they were impressed with the power of Boinc, especially when compared to all the Rental time on Supercomputers they've been paying for. I guess each project knows best what their needs are.

And I wouldn't be at all surprised if they have BOINC do as much of it as possible...but 10 petabytes is a LOT of data. They may still need to rent time or come up with some other resource. Can they get the equivalent of 100K CPU's doing LHC full time from BOINC alone?

---

If you think education is expensive, try ignorance.

The LHC particle smashing data runs may require high bandwidth. There may be portions that require a few GB / sec of data transfer to be worthwhile. There may also be portions that can be farmed out to BOINC. We will have to wait and see.

---

BOINC WIKI

I don't know the cost of renting super compters, and I don't know the costs of bandwith...

But I would think that the costs of renting super computers are higher than the bandwith. And maybe they also could afford the costs of advertising to get more clients? "Help science with your idle computer time, join BOINC!" That would help the other projects as well!

---

Greetings from Norway.
Crunching is fun!

I don't know the cost of renting super compters, and I don't know the costs of bandwith...

But I would think that the costs of renting super computers are higher than the bandwith. And maybe they also could afford the costs of advertising to get more clients? "Help science with your idle computer time, join BOINC!" That would help the other projects as well!

I believe that it was the bandwidth to each individual host, not the bandwidth at their server that needed several GB/S nearly continuosly. Of course the bandwidth at Cern would need to be much higher.

---

BOINC WIKI

The last word we got from one of the BOINC admins (Chrulle?) was that the simplest simulation they could come up with might have a download of 1-5 GB, obviously only practical for people with a lot of disk space and high bandwidth. This is using Geant4 converted to BOINC.

The exciting thing about this is that this would expand the pool of source to all particle labs. But, there are other problems. If someone is interested we can search for the the posts (LHC@Home seems to be down at the moment).

In my case, cable modem, and I can always buy larger disks (heck I think I have 4 300G drives sitting on the floor at the moment ... :)), so, I am interested.

As far as the post processing, the data rate and volume will almost certainly preclude our doing much of that, though I can hope about that too ...

But, I don't think that they will forget about us, the question is can they come up with a reasonable project.

---

The LHC@Home thread has the discussion.

---