|
[Sponsors] |
parallel simulation fails - lscpu shows wrong number of cores |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
May 31, 2022, 17:08 |
parallel simulation fails - lscpu shows wrong number of cores
|
#1 |
Senior Member
Alan w
Join Date: Feb 2021
Posts: 284
Rep Power: 6 |
Hi there,
Recently I got my more capable desktop PC up and running; it has an AMD EPYC 4313P processor with 16 cores. Now the problem is getting programs to run with it. Initially I am try to run OF simulations using 6 cores. Right off the bat I found that I have to add 'oversubscribe' after the mpirun command, as in: mpirun -oversubscribe -np 6 -parallel But the run still failed, with this message: Code:
/*---------------------------------------------------------------------------*\ ========= | \\ / F ield | OpenFOAM: The Open Source CFD Toolbox \\ / O peration | Website: https://openfoam.org \\ / A nd | Version: 8 \\/ M anipulation | \*---------------------------------------------------------------------------*/ Build : 8-1c9b5879390b Exec : snappyHexMesh -dict system/snappyHexMeshDict.1 -parallel -overwrite Date : May 31 2022 Time : 11:31:10 Host : "boffin5-VirtualBox" PID : 2822 I/O : uncollated [0] [0] [0] --> FOAM FATAL ERROR: [0] "/home/boffin5/cfdaero/rad6/system/decomposeParDict" specifies 6 processors but job was started with 2 processors. [0] FOAM parallel run exiting [0] Next, in the ubuntu terminal I entered 'lscpu' to see my system configuration. The output: Code:
boffin5@boffin5-VirtualBox:~$ lscpu Architecture: x86_64 CPU op-mode(s): 32-bit, 64-bit Byte Order: Little Endian Address sizes: 48 bits physical, 48 bits virtual CPU(s): 2 On-line CPU(s) list: 0,1 Thread(s) per core: 1 Core(s) per socket: 2 Socket(s): 1 NUMA node(s): 1 Vendor ID: AuthenticAMD CPU family: 25 Model: 1 Model name: AMD EPYC 7313P 16-Core Processor Stepping: 1 CPU MHz: 3000.002 BogoMIPS: 6000.00 Hypervisor vendor: KVM Virtualization type: full L1d cache: 64 KiB L1i cache: 64 KiB L2 cache: 1 MiB L3 cache: 128 MiB NUMA node0 CPU(s): 0,1 Vulnerability Itlb multihit: Not affected Vulnerability L1tf: Not affected Vulnerability Mds: Not affected Vulnerability Meltdown: Not affected Vulnerability Spec store bypass: Vulnerable Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization Vulnerability Spectre v2: Mitigation; Retpolines, STIBP disabled, RSB fil ling Vulnerability Srbds: Not affected Vulnerability Tsx async abort: Not affected Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtr r pge mca cmov pat pse36 clflush mmx fxsr sse s se2 ht syscall nx fxsr_opt rdtscp lm constant_t sc rep_good nopl nonstop_tsc cpuid extd_apicid tsc_known_freq pni ssse3 cx16 sse4_1 sse4_2 hyp ervisor lahf_lm cmp_legacy cr8_legacy 3dnowpref etch vmmcall fsgsbase arat As a last resort, I looked into the system BIOS, and made this change: Aptio Setup -AMI Advanced CPU Configuration Core Control [seven (7+0)] ...........(this was previously set to [AUTO]) I'm not sure about the 'seven', but that was the biggest option, so I figured that it might correspond to 16 cores (??) I didn't see any other place to specify the number of cores used. After all this, my program still fails, with the same message. Can someone help me untangle this Gordian knot? |
|
June 2, 2022, 10:12 |
|
#2 |
Super Moderator
Alex
Join Date: Jun 2012
Location: Germany
Posts: 3,427
Rep Power: 49 |
In bios, you can only lower the number of active cores. If you want all cores, you have to leave that setting on auto.
Is my guess correct that you are running Windows 10 as a host OS, and then OpenFOAM inside of some virtualized environment? Virtualbox as it seems... Then the problem is with this layer, not on the bios level. Switch core control back to auto, check if all cores are enabled in the host OS, and work from there. I am not exactly the worlds leading expert on virtualization. What I can say is if you don't want to leave a lot of performance on the table for OpenFOAM, don't use Virtualbox. There is WSL as an alternative, but from the benchmarks I have seen, it is also significantly slower than bare metal. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
parallel run in foam extend | CRI_CFD | OpenFOAM CC Toolkits for Fluid-Structure Interaction | 5 | March 28, 2024 09:36 |
[Other] Can't Shake Erros: patch type 'patch' not constraint type 'empty' | BrendaEM | OpenFOAM Meshing & Mesh Conversion | 12 | April 3, 2022 19:32 |
Compressor Simulation using rhoPimpleDyMFoam | Jetfire | OpenFOAM Running, Solving & CFD | 107 | December 9, 2014 14:38 |
foam-extend_3.1 decompose and pyfoam warning | shipman | OpenFOAM | 3 | July 24, 2014 09:14 |
[Commercial meshers] Trimmed cell and embedded refinement mesh conversion issues | michele | OpenFOAM Meshing & Mesh Conversion | 2 | July 15, 2005 05:15 |