21) Message boards : ATLAS application : This is not SLC6, need to run with Singularity (Message 35936)
Posted 16 Jul 2018 by AuxRx
Post:
One of your computers is configured to use 4 cores for ATLAS native. Generally speaking, the lower the CPU cores per ATLAS task, the higher the efficiency of that task (if you define it as (CPU time)/(task runtime * number of cores), i.e. it is more efficient to run concurrently 4 1-core tasks than 1 4-core task. So to increase efficiency, the first thing i would do is to reduce the number of cores per ATLAS task from 4 to 1 IF you have enough RAM for that (also running 2 2-core tasks would be more efficient than your current setup and propably the better choice considering your 8GB RAM limitation).


That's news to me. How could the efficiency of the compute improve with fewer cores? Are you refering to eliminating idle time connected with starting and stopping tasks? What is your metric?

I am familiar with the diminishing returns of Hyper Threading (or SMT in general) and issues with the credit system, but those are different arguments.
22) Message boards : Sixtrack Application : AVX Sixtrack version (Message 35778)
Posted 5 Jul 2018 by AuxRx
Post:
Any chance we can re-open this issue? I've been stuck with SSE2 instead of AVX for too long. At this point I doubt this is going to remedy itself. The algorithm that is supposed to re-evaluate my system is not doing its job. It might even be the culprit, since I was getting AVX initially.
23) Message boards : Number crunching : The Server (Message 35739)
Posted 2 Jul 2018 by AuxRx
Post:
Any news from the team yet? Anything users have to do or know?
24) Questions and Answers : Sixtrack : no more task ! (Message 35650)
Posted 27 Jun 2018 by AuxRx
Post:
There are no tasks available at the moment. You can check the server status tab for more up to date information.
25) Message boards : Sixtrack Application : Inconclusive results (Message 34757)
Posted 27 Mar 2018 by AuxRx
Post:
We've had similar issues during the last few database troubles. I expect it'll solve itself in a couple of days.
26) Message boards : Sixtrack Application : AVX Sixtrack version (Message 34724)
Posted 25 Mar 2018 by AuxRx
Post:
My system stopped receiving AVX tasks on Thursday, is consistently getting SSE2 instead.
27) Message boards : Number crunching : Hyperthreading vice no hyperthreading for VM tasks (Message 34666)
Posted 15 Mar 2018 by AuxRx
Post:
Any advice on setting up a cache? Did you use proprietary software or which solution did you go with? How do you deal with reboots?
28) Questions and Answers : Unix/Linux : ATLAS native Linux app (Message 34525)
Posted 2 Mar 2018 by AuxRx
Post:
Suspending and resuming tasks hasn't worked (reliably?) with the native app for quite sometime.
29) Message boards : ATLAS application : ATLAS native app (Message 34508)
Posted 28 Feb 2018 by AuxRx
Post:
Finally received native_mt! Thank you for your support, when I began questioning my sanity.

As to what finally did the trick, I have no idea. The last thing I changed was resetting the "run test applications" in my account preferences and rebooting. Maybe I'll try reproducing the issue, but atm I'll try running it for a couple of days.
30) Message boards : ATLAS application : ATLAS native app (Message 34500)
Posted 27 Feb 2018 by AuxRx
Post:
Thank you, but no change unfortunately. I either get vbox tasks or no tasks.

I was able to verify that Singularity is working with an example file.
31) Message boards : ATLAS application : ATLAS native app (Message 34497)
Posted 27 Feb 2018 by AuxRx
Post:
Thank you very much for your detailed reply!

Concerning CVMFS:
I went the exact same route. Well, except I modified fuse.conf and auto.master after the first failure.
    * Is it safe to assume that modifications to fuse.conf and auto.master are not necessary?

    * Did sudo cvmfs_config chksetup complain about geo-API (seems to be related to GRID)?



Singularity:
I guess I'll further focus on Singularity.

    * How can I check whether Singularity (by itself) is working properly?


Again, thank you!

*** EDIT: I tried building Singularity and will report back in case I receive native ATLAS.

32) Message boards : ATLAS application : ATLAS native app (Message 34493)
Posted 27 Feb 2018 by AuxRx
Post:
A simple question: Is AMD-V or VT-X enabled in Bios?

Yes, and confirmed working with Virtualbox.

Have Proxy ="DIRECT" in the declaration for SL69.

Can I ask why? The documentation suggests DIRECT (without quotation marks).
How did you make the call not to use the proxies CERN included in the script at /boinc_conf/?

Frankly I don't even know what to try, ask or look up next.
33) Message boards : ATLAS application : ATLAS native app (Message 34491)
Posted 27 Feb 2018 by AuxRx
Post:
After a longer break I thought I would give LHC@Home another try, and started with a clean install of Ubuntu 16.04.3 LTS. Previously, I had no issues with the native app, despite being new to Linux. However, the intermittent issues several users reported before now have caught up to me.

To help me, I am using the script at
http://atlasathome.cern.ch/boinc_conf/

as well as the information provided in both
https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4502#33029
https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4396&postid=34346

At last I resorted to the CVMFS documentation.
https://cernvm.cern.ch/portal/filesystem/quickstart

After several retries I still can't get the native app to work.
Removing vbox or forcing "dont_use_vbox" simply produces a message that vbox isn't installed.
cvmfs_config probe, status, stat *repository*.cern.ch and chksetup all work.
/scratch/cvmfs contains what I assume to be a CERN network share.

These are my observations:
1) There is some ambiguity about default.local in the sources I mentioned above. The script at /boinc_conf/ suggests using the *internal* CERN proxies. Others suggest replacing the proxies with "DIRECT" (with quotation marks) or DIRECT (w/o quotation marks). Is using the CERN proxies advisable? What is the correct value/syntax?

All three alternatives "work" i.e. don't obviously break anything. cvmfs_config stat atlas.cern.ch simply reports the different settings.

2) cvmfs_config chksetup is complaining about Geo-API not working, but I remember this being normal even when CVMFS worked. Does this need to be addressed?

3) The documentation uses cvmfs-config-default and cvmfs_config setup. No further editing of fuse.conf or auto.master is required. However "/misc /etc/auto.misc" and "/net -hosts" in auto.master aren't uncommented i.e. put to use. Is this necessary for full functionality?

4) Why not use the script at /boinc_conf/? Cause it uses yum and selinux that I don't use on Ubuntu. It didn't work anyway.

5) It's difficult to diagnose the issue, because there is no reliable supply of ATLAS tasks. I can never tell, whether there are no tasks or I am not sent any.

I am using a fresh install of Ubuntu 16.04.3 LTS
BOINC 7.6.31 from the ubuntu rep
VirtualBox 5.2.x from the Oracle PPA
Singularity-container 2.4.1-2 from Neurodebian rep (Ubuntu 17.10 would still use 2.3.1?)
CVMFS 2.4.3 from the CERN PPA
34) Message boards : Number crunching : Message Board Issues ??? (Message 34357)
Posted 10 Feb 2018 by AuxRx
Post:
This is normal behavior atm.

Specifically, when I opened the thread to read her message, her message was the only post on the thread


Not sure if this a question, but replying to a thread and replying to a post is different. Replying to a post will only show the related post underneath the text area while you type.

So it looks to me that I just happened to be posting during a time when the MB server may have been having some transient issues.


As you suspected, some one answered while you were drafting your answer and that screwed up your post. You might even find the server didn't accept your post, which has happened to me before.
35) Message boards : Sixtrack Application : "no tasks are available for Sixtrack", over 1 million visible (Message 34227)
Posted 1 Feb 2018 by AuxRx
Post:
I have a very different perspective. SixTrack is a very active sub project, because it's system requirements are generally low and downloads are small. You'll find systems on Sixtrack that no other LHC project could support.

Not requiring VBOX is a major part of that. Vbox installs are troublesome still. New users are often turned away from LHC because they don't want to deal with the checklist -- and they shouldn't have to. ATLAS native App is no solution because it adds several layers of difficulty. Also, the native app does very little for users even if they are on linux. Major complaints remain download size and system requirements. My system, at least, received the same size downloads in addition to CVMFS traffic. If LHC can use a cache that's great, but doesn't help users with limited internet traffic. And I didn't see improvements in system usage or stability (pausing tasks specifically).

I fear users will be excluded if changes are made that profit a small minority.
36) Message boards : Sixtrack Application : Transfer issues (Message 34119)
Posted 27 Jan 2018 by AuxRx
Post:
Long validation times don't bother me personally. But it would be prudent to control job creation by staggering job creation. AFAIK over-filling the queue with large batches does not speed up production. Inf act, at work we try to achieve the opposite in accordance with Best Practice, Lean Management principles.

Can job creation be initiated automatically and staggered in smaller batches? Could job creation be rescheduled so staff is available to deal with hiccups?
37) Message boards : Sixtrack Application : Transfer issues (Message 34112)
Posted 27 Jan 2018 by AuxRx
Post:
I've run out of work this morning. My system has been asigned new tasks, but the download is delayed again and again.

Not worth it, shutting down for the weekend.
38) Message boards : Number crunching : Error while computing/too many errors (Message 34106)
Posted 26 Jan 2018 by AuxRx
Post:
... same comment really, if the machine was off or busy or whatever, it trashes a work unit, I doubt that.


Sorry, but that's the way it is. It's all about Condor connections.
39) Questions and Answers : Unix/Linux : LHC@Home on ARM v7 and v8 processors (Message 34091)
Posted 26 Jan 2018 by AuxRx
Post:
At this time only SIXTRACKTEST supports ARM. For more information see: https://lhcathome.cern.ch/lhcathome/forum_thread.php?id=4296
40) Message boards : News : Thanks for supporting SixTrack at LHC@Home and updates (Message 34062)
Posted 24 Jan 2018 by AuxRx
Post:
Thank you very much indeed! Great summary, exactly what I have been looking for!

I've been thinking about another great addition (sorry ...). On the ATLAS site we had progress bars for individual simulations (i.e. all of workspace1_hl13_collision_scan_*). Even with the limited amount of work I can contribute, it really feels like an accomplishment to see those progress bars running down. Otherwise the individual volunteers progress is too slow and the transition to a new simulation often gradual and uneventful.

https://lhcathome.cern.ch/ATLAS/index.php

Again, thank you for the update!


Previous 20 · Next 20


©2024 CERN