LHC@home

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

LHC has been planned and budgeted for YEARS.

They must have had a plan....

Anyone?

---

They would have run the Sixtrack program on a bank of computers.

When doing an initial design, you don't need to run a million cycles to pick out how big of a magnet you need to bend a particle beam around a huge circle.
Running a million cycles lets you fiddle with the controls under a variety of conditions.

In the old days, they just built the cyclotron, and then fiddled with the controls.

---

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

They would have run the Sixtrack program on a bank of computers.

When doing an initial design, you don't need to run a million cycles to pick out how big of a magnet you need to bend a particle beam around a huge circle.
Running a million cycles lets you fiddle with the controls under a variety of conditions.

In the old days, they just built the cyclotron, and then fiddled with the controls.

So how much did they save by not having to buy this bank of computers?

---

They would have run the Sixtrack program on a bank of computers.

When doing an initial design, you don't need to run a million cycles to pick out how big of a magnet you need to bend a particle beam around a huge circle.
Running a million cycles lets you fiddle with the controls under a variety of conditions.

In the old days, they just built the cyclotron, and then fiddled with the controls.

So how much did they save by not having to buy this bank of computers?

Cern already has banks of computers. Instead of running a work unit for 1 million cycles, you could run 1000 work units for 1000 cycles to weed out the blatantly unstable control situations.

After all, there's real world experiments.

---

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

They would have run the Sixtrack program on a bank of computers.

When doing an initial design, you don't need to run a million cycles to pick out how big of a magnet you need to bend a particle beam around a huge circle.
Running a million cycles lets you fiddle with the controls under a variety of conditions.

In the old days, they just built the cyclotron, and then fiddled with the controls.

So how much did they save by not having to buy this bank of computers?

I think I read somewhere an article by Chrulle that said that during the extended beta (5000 users) it was the equivalent of having 200 * P4 2.8GHz HT machines running 24/7.