LHC@home

I suspend my participation unless someone convince me that the computations on already collected data are still useful to find something in the new physics.

We have never calculated here any collected data. Everything has been simulations for new and future experiments.

---

That's right. BOINC users only ran simulations, never crunched real data.

One can olny hope that our simulations are somehow useful at all !?
But i think if they weren't LHC would not maintain the infrastructure for BOINC.
On the other hand, LHCb dropped out for monetary reasons (maybe CMS too?).

Many uncertaincies...

---

Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us

So sad....

---

Sooo much data is collected that it would be impossible to analyse it all. The simulations we run give the scientists pointers as to where to look in that data.

Just for my personal understanding...
If there is so much data being collected, why is exactly that data not being distributed to the BOINC crunchers?
Wouldn't that make more sense? When the BOINC computing power is equivalent to that of a WLCG Tier-2 site, why doesn't CERN use this power to get real results?

There is a current example for what computer simulations can be worth: The Ocean Cleanup Project.
They ran extensive simulations too, but the real product doesn't work the way it's supposed to and they wonder why that is (quote: In simulations this problem never occured.).

Don't get me wrong. I totally understand that simulations are important to figure out where to look in the real data, but why don't make use of both?

Regards!

---

Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us

I think I've heard that the collected data is not allowed to be analyzed on an open system like Boinc. For security and data integrity reasons the analyzing must be done inside Cerns computing grid. Just look at the graphics here: https://lhcathome.cern.ch/lhcathome/atlas_job.php below the lower graph you can see a list of computing centers running the Atlas simulations. Boinc is quite a small part of that list, although we are in top 5 performers. I assume that those centers are also used for collected data analyzis? Does anybody have actual info on that?

---

I think I've heard that the collected data is not allowed to be analyzed on an open system like Boinc. For security and data integrity reasons the analyzing must be done inside Cerns computing grid.

That would make perfect sense to me...unfortunately...

I'm crunching LHC tasks for more than 4 years now and i'm starting to wonder if my contributions have any positive impact on the science being done at CERN.

There are other projects where actual data is being distributed for crunching and i'm considering switching there, especially giving the fact that the LHC has found nothing new in the last 10 years. It only proved the existence of the Higgs-Boson and that neutrinos have a mass. That's it. Nothing new, what wasn't predicted by theories which are 50 years old and older. No extra dimensions, nothing on dark matter, no new elementary particles, no supersymmetry...I know that's a little bit exaggerated, but anyway.

So i have to join Bernard in his wish to be convinced that we do something scientifically useful and to stick with LHC@home.

Greetings!

---

Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us

Albert Einstein's gravition waves in theory needed 100 years to be in practice analyse at Ligo.
Science is nothing for only some years. Cern is doing many projects for understanding of the universe.
Every Volunteer need to ask himself, what can be a small part he can do.

Every Volunteer need to ask himself, what can be a small part he can do.

And that's exactly my question here...am i REALLY doing something useful that actually helps CERN?
Is there any example where a LHC simulation done by a cruncher helped with analyzing real data?

---

Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us

Is GRIDCOIN really?

Since Gridcoin is very off topic here, i send you a private message...

---

Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us

I will try to give a short and simplified answer.

To do the science you need to compare real experimental data with the theory. Data about collisions can be produced from the experiment or via simulation (the equations from theories and models). Both sets of data follow the same analysis chain. That is taking the data from numerical values of sensors (think of a raw image from a 100 mega pixel camera) and then looking for lines and curves. People tend to talk a great deal about the amount of data coming from the detectors and the fact that we need at least the same amount of simulated data is often overlooked. Simulation fits with volunteer computing as it is CPU intensive rather than data intensive.

The SixTrack application, which is slightly different, runs simulations of the beam itself. These simulations are used for the purpose of design, development and operations. This could be run 3 scenarios, future upgrades or future accelerators.

This all means that even if the LHC is shutdown, simulations continue. Internally we re-purpose our trigger farms (~30K cores underground at each detector) to run simulations during this time.

Laurence,

thank you very much for your explaination, now i know that i'm really contributing something useful for the science CERN does. LHC@home is my favorite BOINC project and i only reluctantly would have left this project. Also thanks to CERN IT for maintaining the infrastructure for BOINC, i will joyfully keep on crunching!

Yours, djoser.

---

Why mine when you can research? - GRIDCOIN - Real cryptocurrency without wasting hashes! https://gridcoin.us

Just to add to what Laurence wrote - the data which comes from the detectors is first processed by dedicated powerful computing farms inside CERN, then the resulting data is distributed around the world for analysis or further processing. This first processing step is very data intensive and time-critical which doesn't suit volunteer computing very well. Simulation on the other hand is less data-intensive and does not demand immediate processing so this is what we run on LHC@Home.

As Laurence said, it is often not appreciated how much resources are needed to produce the simulations needed to analyse the real data. In ATLAS around 2/3 of the entire computing grid taken up with running simulations and it's still not enough, so what you are doing here on LHC@Home really adds to the results.