LHC@home

The application LHCb run on GPU is the trigger system HLT 1, which reduces data rate from 40Tbps to 1-2Tbps. I don't think you have 40Tbps internet connection at home, so we certainly won't participate in running this application.

For optimal performance and hardware-utilisation, it is crucial to share the particle state in-place between the codes SixTrackLib (used for tracking single particles following collisions) and PyHEADTAIL (used for simulating macro-particle beam dynamics). This helps to avoid the memory and run-time costs of maintaining two copies on the GPU. The current implementation allows this by virtue of implicit context sharing, enabling seamless hand-off of control over the shared state between the two code-bases. After a first proof-of-concept implementation was created at an E4-NVIDIA hackathon in April 2019, the solution was refined and the API of the libraries was adapted to support this mode of operation.

Work carried out within this project made it possible for PyHEADTAIL to rely on SixTrackLib for high-performance tracking on GPUs, resulting in performance improvements of up to two-to-three orders of magnitude compared to state-of-the-art single-threaded CPU-based code. We also exposed SixTrackLib to new applications and use cases for particle tracking, which led to several improvements and bug fixes.

As some of you argued, we are focusing on using OpenCL 1.2 because it allows to run on AMD, Intel and Nvidia GPUs, although is less advanced than other options (OpenCL 1.2 dates back to 2011).

[quote]I want to finish this one
67% after 67hs

You may also have a look in the console with ALT-F1 for the job description and ALT-F2 for progress.

2.73295e+18 pb +- ( 6.02271e+17 pb = 22.0374 %) 62802000 ( 628025526 → 99.9% )
Integration time : (1d 16h 57m 52s elapsed / 827d 12h 52m 33s left ) [15:42:29]
Poincare::Poincare(): Inaccurate rotation {
a      = (-0.91078,-0.399788,0.101433)
b      = (0,0,1)
a'      = (0.317924,-0.918998,0.233165)→ rel.dev. (inf,-inf, -0.76635)
m_ct =0.101433

ecc, ecc

You have to use 'Show VM Console' to see what's going on.

Tasks: 110 total, 2 running, 71 sleeping, 0 stopped, 0 zombie
%CPU(s): 96.6 us, 2.8 sy, 0.0 ni, 0.0 id, 0.2 wa, 0.0 hi, 0.4 hi, 0.0 st
KiB Mem: 605712 total, 64660 free, 312404 used, 228648 buff/cache

5504 boinc   20   0   473896   39124   5736   R   99.2   6.5   2213:28  Sherpa
5504 boinc   20   0   473896   40692   7304   R   98.8   6.7   2213:43  Sherpa

Should we automatically abort them? Easy to do it by bash script.

It's your decision.

We are still in the process of developing SixTrackLib, validating the physics and tackle the integration into the main SixTrack. Our time scale is still probably a year or two to expose the functionality into LHC@Home, but we already performing simulations using SixTrackLib in GPU for specific problems.

Then how come there are projects which only use ATI but not Nvidea?

I can imagine that sixtrack would be the one that would make a GPU application. The challenge for the project team is finding the time to port the application, it has been asked for quite a few times.

At first i want to speek from the computer market of GPU
The most selled GPU are NVIDIA GTX...

Ok, i've installed the latest version of VirtualBox and the latest extensions.

2019-10-21 17:05:28 (5416): Guest Log: HITS file was successfully produced

If you are interested in a better task monitoring those extensions must be installed.
See Yeti's checklist:
https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4161&postid=29359

The task finished and got credits - that's BOINC's point of view.
But you may be aware that from the scientific point of view the task didn't succeed:

2019-10-12 05:27:54 (12056): Guest Log: No HITS file was produced

Your logfile shows that the task started/paused several times and couldn't successfully write it's snapshot

Wu arrived quickly to 99%, then slowed down and now, after 22hs is at 100%, but still crunching.

Do you have an idea when they might finish?
So far they reached 100% in 1-2hrs but the FLOPS estimate must be way wrong since they are still going after a day or so.

First of all the German Rechenkraft.net team would like to thank Ben Segal, Ian McIntosh, David Cameron and all the others involved for organizing both the ATLAS@home and the LHC@home booths which our team visited with 13 people on Saturday, 14th.

Of course, due to so many people, waiting times to enter the underground sites were ridiculously long: If I remember correctly, on the first day we waited almost 4 hours after entry at 9 am to only see the ATLAS detector for around 20 minutes. And this was located nearby, directly at the main entrance (resulting in people rushing there first, of course).

A nice thing for the future would be a yellow VIP wristband for volunteer computing participants which should allow us to directly proceed to the underground sites - omitting the waiting queues of course... ;-)

You needed a smart phone to have any chance. I am complaining to organisers, why was there NO MAP!

During the CERN Open Days 2019, we will have a small LHC@home stand as part of the IT activities in building 504 near the Data Centre.