Message boards :
ATLAS application :
2000 Events Threadripper 3995WX
Message board moderation
Previous · 1 · 2
| Author | Message |
|---|---|
 Send message Joined: 15 Jun 08 Posts: 2413 Credit: 226,606,378 RAC: 130,379    |
Read the kernel docs and the systemd docs. |
|
Send message Joined: 2 May 07 Posts: 2104 Credit: 159,819,191 RAC: 123,837 
|
This machine is running a 1000 events job on 4 cores and is already 26 hours busy and another 7 hours to go. Got three 100 event Tasks: Laufzeit 24 min. 40 sek. CPU Zeit 2 min. 41 sek. Is this useful? |
|
Send message Joined: 3 Nov 12 Posts: 36 Credit: 117,893,930 RAC: 127,327
|
What about reactivation of the "Atlas long simulation" for the 24/7 systems? It was something I liked. |
|
Send message Joined: 2 May 07 Posts: 2104 Credit: 159,819,191 RAC: 123,837
|
+1 |
|
Send message Joined: 14 Sep 08 Posts: 43 Credit: 52,026,498 RAC: 140,592
|
Yep, ATLAS long seems to be the perfect answer to break this dilemma. Another good part about ATLAS long is that it's a separate app. Now I can put in a different nthreads in app_config to throw more cores at the bigger problem without worrying about having 8 cores wait for 30 minutes just to do 10 minute of compute... |
|
Send message Joined: 4 Sep 22 Posts: 57 Credit: 8,489,207 RAC: 17,667   
|
Read the kernel docs and the systemd docs. You remind me of the guy in the bad old days of DOS and no internet, who steadfastly refused to document his code. His rationale? "People who can't read code have no business running a computer." |
|
Send message Joined: 14 Sep 08 Posts: 43 Credit: 52,026,498 RAC: 140,592
|
Overloading a system most of the time just to make up the idle time is not a real solution anyway. Fixing the workload to not idle is generally the best solution. The 20 minute time is with local cvmfs installation using cloudflare CDN. If during the time, the workload is pulling lots of data down from cvmfs and limited by bandwidth or disk, then I can more or less understand the behavior. However, it's not doing anything remotely intensive with network or disk either. The problem is likely either how the setup pulls the necessary data, or how fast cvmfs can serve the data, or both. But hey, this is the only program on my computer that's running python 2.7 which has reached EOL 4 years ago. This also means it's definitely not using basic optimizations like asyncio to parallelize IO requests. I doubt much care for performance is put into the setup phase, given the actual calculation is likely done by some C/C++ library. |
|
Send message Joined: 2 May 07 Posts: 2104 Credit: 159,819,191 RAC: 123,837
|
CVMFS and Parrot is the alternate for CVMFS and Fuse. Would be good, to install CVMFS on the own network. Don't know under Win11pro or CentOS9? This answer is from Bing KI: Parrots and CernVM-FS: While parrots themselves don’t directly interact with CernVM-FS, the term “Parrot” here refers to the Parrot connector used as an alternative to the Fuse module |
©2024 CERN