LHC@home

I am running vLHC@home on three PCs, two Linux 32-bit and one Windows 8.1 64-bit. They run all without a glitch on Virtual Box 4.3.12 and are all validated. ATLAS@home, running on the Windows PC, is more subject to validation errors due to the different architectures. I am running it on a AMD A10-6700 together with Einstein@home and SETI@home besides vLHC@home. Einstein and SETI make use of a Radeon HD 8670D graphic GPU. In SETI I am using the Lunatics installer.
The critical point of ATLAS@home is its memory load, 2 GB per unit. Since I have a 8 GB RAM I am letting one ATLAS@home task and 2 vLHC tasks to run using Virtual Box, leaving space also to another program on the fourth core of the AMD APU. Buth SETI GPU and Einstein GPU tasks use very little CPU (0.04 SETI, 0.5 Einstein). Only one GPU program can run at a given moment. Einstein always grabs the GPU from SETI. SETI sends me 10 or 11 GPU tasks but they are processed one at the time.
Tullio

I tried vLHC and ATLAS on 3 different machines, with several distros and distro versions, and none worked.

I've started vLHC@home in November 2010 when it was called Test4Theory@home. I was invited by Ben Segal, who had read a post of mine in another project where I spoke about my experience with a Solaris Virtual machine I had installed as a test of Virtual Box on a Linux box. The Alpha phase of T4T was very troubled, but slowly it emerged as a very reliable project and now it gives me no problem both on Linux and Windows.
In my opinion the ATLAS developers, which include a charming Chinese girl, were too much in a hurry. Also, they started downloading a number of tasks equal to the number of cores in the host, 4 in my case. But since ATLAS tasks require 2 GB each, this causes a problem in RAM. I have only 8 GB RAM on my Windows PC, and it can run either 2 ATLAS tasks and no vLHC task, or one ATLAS task and 2 vLHC tasks. vLHC started allowing only one task/host, now it allows 2.
I've suggested the same policy on the ATLAS message boards, but people having more cores in their CPU protested. It seem that most of them have 16 GB RAM or even more. I could go up to 32 GB RAM on my HP 500-152ea system but I still remember the time I had a 2.5 MB (not GB) RAM on my AT&T Unix PC. a.k.a. PC703 or Safari. It has a 40 MB hard disk and it still works running UNIX, no X-Windows.
Tullio

That's one of the biggest drawback of VM; it requires a huge amount of RAM because 2 OS are running!!!
It is too heavy for common PC/Mac users which have "only" 4GB or 8GB of RAM installed (as you said tullio).
Further more, android platform (smartphone, tablet,...) can't run VM.
My advice is to stop thinking about VM for this project and to focus on GPU porting (Except if you working for a virtualization company).
Because with the app_config.xml BOINC feature, you can run several tasks on the same GPU.
Personnaly, all my GPUs run 2 tasks at the same time for all projects I participate to. And I have several GPUs on my PCs which represents a big efficient and potential ressource that LHC@home can't use.

My quote is: "I'm tired of computing 10h WUs while only 1h would be enough."

---

WU's on first

Well, I am only a pensioner and don't work anymore. Today I was interested in trying the new release of SuSE Linux, 13.2, in its 64-bit version so it could run also ATLAS@home. So I installed it as a Virtual Machine on my SuSE Linux 13.1 32-bit where I have a lot of stuff also compiled by myself. Then I installed the last version of Virtual Box 64-bit on it, and since the one downloaded from SuSE did not work, I compiled it. But first I had to download the gcc compiler, the kernel sources and the Make tool. Then I installed BOINC 4.2.42 but had to register to vLHC@home using the line of command BOINC client, since the Boincmanager did not work on the new release, as it often happens. Releases of some tools not always work on new releases of an operating system.
So time has little importance for me, I don't like systems which need only point and click. Once, while working, I compiled the GNU GCC 2.2 compiler using the UNIX System V compiler, and using this GCC, which is an ANSI compiler, I compiled from the sources TeX 3.14 and GRASS, a land surveying and mapping program developed by the US Engineering Corps. So I passed a Sunday afternoon and can go to bed.
So I haave created another second tier virtual computer crunching vLHC@home.
Tullio
How come you don't crunch the more interesting CERN programs like vLHC@home and ATLAS@home?

Talking about VM is off topic on this thread (LHC on GPU ?). Choose or create another thread for VM topic.

---

WU's on first

OK, let us stop here.
Tullio

Looking at it; but will be several months at least.
See my next status report coming soon. Eric.

To bring this thread back on topic. I have had a look through the news posts but I am not able to find anything relating to running LHC on GPU's. Is anybody able to post link to the status report in question?

---

Have A Crunching Good day