Development for LHC@home

As we now have multiple applications, some of you have requested that we remove the restriction that limits the number running tasks per host to one. This restriction has was put in place so as not to overload machines while developing. As everyone in this development project is (or should be) an advanced user, we assume that everyone knows how to adjust their preferences to limit tasks on the client if needed. This update will be done tomorrow morning at around 10am CET (8am GMT, 1461052800 UTC).

That's a great idea.
If it causes Furious Confusion amongst the community, I'm sure I/we could cobble up an FAQ showing How To Get What You Want.

One thing that worries me though - you'll have to replace Ivan's cute logo with something more generically CERN!

As a reminder, if you are using a modern enough version of BOINC, you can put various configuration commands in a file called app_config.xml in your BOINC project directory. e.g. on my Linux machine for this project that is /home/ivan/BOINC/projects/lhcathomedev.cern.ch_vLHCathome-dev/app_config.xml.

The particular commands to limit the number of concurrent instances (to 1 in this case) for CMS is:

<app_config>

	<app>

		<name>CMS</name>

		<max_concurrent>1</max_concurrent>

	</app>

</app_config>

Other application names include:
theory64 for Theory Simulations 64 bit
vboxwrapper for Theory Simulations

More details can be found in the BOINC application configuration documentation.

---

One thing that worries me though - you'll have to replace Ivan's cute logo with something more generically CERN!

I did try to come up with something combining the CMS and LHCb logos when we started to diversify, but the different aspect ratios of the two made a combination difficult without emphasising one over the other.

---

Other application names include:
theory64 for Theory Simulations 64 bit
vboxwrapper for Theory Simulations

Are you sure?

I thought
CMS
LHCb
Theory
ATLAS

are the app names for this project.

As we now have multiple applications, some of you have requested that we remove the restriction that limits the number running tasks per host to one. This restriction has was put in place so as not to overload machines while developing. As everyone in this development project is (or should be) an advanced user, we assume that everyone knows how to adjust their preferences to limit tasks on the client if needed.

Well most of the time anyway....but once in a while some jump over the fence or dig under the fence to get in here

BUT my preference is usually jump right in a try all of the versions

.....as long as it is Cern or Einstein GPU's

(85 degrees here so my fleet of GeForce cards are a bit warm today)

---

Mad Scientist For Life

Other application names include:
theory64 for Theory Simulations 64 bit
vboxwrapper for Theory Simulations

That are the names for the production project, not for this dev-project. The names are what Rasputin wrote:

ATLAS
CMS
LHCb
Theory (without 64)

Thanks for reminding me. I have replaced the logo with the vLHC one. Feel free to hack it and make it unique for this dev project. I was thinking that a hammer hitting the computer would be appropriate.


EDIT: And for those who are sad to see the CMS logo go, don't worry, it has been promoted to the front page.

http://lhcathome.web.cern.ch/projects/cms

The flood gates have been opened.

It downloaded 3 more CMS tasks, but it is not running them.
I limitted the number to 1.(max_concurrent)
So far, so good.

My question, why is it even downloading tasks, as my work buffer is set to 0.03 (0) days?

edit: Work requested

<work_req_seconds>6998.400000</work_req_seconds>

<cpu_req_secs>6998.400000</cpu_req_secs>

form the scheduler_request.

I am running two vLHC@home tasks and one vlHCathome-dev on my Windows 10 PC. They all run CMS tasks. On my 64-bit Linux box I have installed vLHCathome-dev and a very short Atlas task has run and validated. Only the 32-bit Linux box is still running good old vLHC@home.
Tullio

The flood gates have been opened.

... and I was almost drowned.

Maybe it's a better idea to limit the number of tasks project-wise to the number of reported threads.

The first 6 CMS-tasks returned all due to: 1 jobs failed out of the last 1 jobs. Shutting down!,
cause I had to reset my frozen machine.

You can limit the total tasks running at the same time for this project by adding:

<project_max_concurrent>x</project_max_concurrent>

to your app_config.xml.

"x" is the number of tasks running simultaniously.

(edit: boinc ver 7.4.9 and up)

Yes, I knew and even advised that to other users at several projects, but I was kept by surprise suddenly running 8 CMS-tasks on my 16GB machine.
That's why it was frozen inaccessible and I had to soft-reset the machine.
Meanwhile created an app_config.xml:

<app_config>
<project_max_concurrent>4</project_max_concurrent>
<app>
<name>ATLAS</name>
<max_concurrent>1</max_concurrent>
</app>
<app>
<name>CMS</name>
<max_concurrent>2</max_concurrent>
</app>
<app>
<name>LHCb</name>
<max_concurrent>1</max_concurrent>
</app>
<app>
<name>Theory</name>
<max_concurrent>2</max_concurrent>
</app>
</app_config>

Otherwise, I think on a development and testing project it's not needed to have a cache of tasks over the maximum available threads.

Other application names include:
theory64 for Theory Simulations 64 bit
vboxwrapper for Theory Simulations

That are the names for the production project, not for this dev-project. The names are what Rasputin wrote:

ATLAS
CMS
LHCb
Theory (without 64)

Ah, OK; sorry for the confusion.

---

The flood gates have been opened.

No problems with any extra tasks running but 1 machine had been running 2 at a time and getting through them in about 7 mins due to...

2016-04-19 09:00:39 (4576): Guest Log: [INFO] Theory application starting. Check log files.
2016-04-19 09:06:29 (4576): Guest Log: [ERROR] App is not supported. Shutting down!

Had been doing that for a few hours before the gates opened and so with estimated run times that low it downloaded over 800 !

Have put the limit to 5 tasks in progress.

Have put the limit to 5 tasks in progress.

Yep, thats about the limit for a 16GB host.

I can confirm functionality of CP's app_config. It didn't like my own version as I didn't have the line <project_max_concurrent>4</project_max_concurrent> but it's happy now except that, not having gotten ATLAS or CMS test jobs yet, it complains about that but it's good-to-go when those do arrive.

Have put the limit to 5 tasks in progress.

Yep, thats about the limit for a 16GB host.

Aww, but what about my 128 GB host (with 20 cores...)?

---