-
Notifications
You must be signed in to change notification settings - Fork 507
The usage of CPU for lammps with deepmd potential is only about 30% #318
Replies: 1 comment · 4 replies
-
Please use the MPI package that is used to compile the LAMMPS, i.e. |
Beta Was this translation helpful? Give feedback.
All reactions
-
Thanks very much for your reply. Now I use this command to perform MD: The usage is still low!! PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND Thu Dec 31 15:27:31 2020 +-----------------------------------------------------------------------------+ |
Beta Was this translation helpful? Give feedback.
All reactions
-
Do you use CPU version or GPU version? Why did you see it using |
Beta Was this translation helpful? Give feedback.
All reactions
-
Everything I used is from "deepmd-kit-1.1.0-gpu-Linux-x86_64.sh." |
Beta Was this translation helpful? Give feedback.
All reactions
-
Well, there is no problem after deepmd-kit is upgraded to 1.3.1. Everything looks OK ! |
Beta Was this translation helpful? Give feedback.
-
Hello, Dear DeepMD users,
I try to perform MD simulations by CPU based lammps with trained deepmd potential.
Each node of the HPC has 54 CPU processors and 12 K80 GPU processors.
The command for performing MD is
mpirun -np 12 lmp -in in.lmp
The usage of each CPU processor is only about 30%.
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
36426 chenwei 20 0 0.271t 1.503g 0.996g S 34.4 1.2 3:17.74 lmp
36423 chenwei 20 0 0.271t 1.500g 0.996g S 33.4 1.2 3:17.04 lmp
36424 chenwei 20 0 0.271t 1.504g 0.996g S 32.1 1.2 3:21.26 lmp
36427 chenwei 20 0 0.271t 1.504g 0.996g S 32.1 1.2 3:16.77 lmp
36418 chenwei 20 0 0.271t 1.505g 0.996g S 31.8 1.2 3:19.87 lmp
36421 chenwei 20 0 0.271t 1.501g 0.996g S 31.5 1.2 3:16.69 lmp
36428 chenwei 20 0 0.271t 1.503g 0.996g S 31.1 1.2 3:18.11 lmp
36422 chenwei 20 0 0.271t 1.503g 0.996g S 30.1 1.2 3:22.58 lmp
36429 chenwei 20 0 0.271t 1.502g 0.996g S 29.8 1.2 3:19.35 lmp
36420 chenwei 20 0 0.271t 1.506g 0.996g S 29.5 1.2 3:16.88 lmp
36419 chenwei 20 0 0.271t 1.503g 0.996g S 28.8 1.2 3:16.52 lmp
36425 chenwei 20 0 0.271t 1.506g 0.996g S 28.8 1.2 3:15.43 lmp
Then I check the usage of GPU by "nvidia-smi". I found each card has 12 processes.
Wed Dec 30 09:19:07 2020
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 410.79 Driver Version: 410.79 CUDA Version: 10.0 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 Tesla K80 Off | 00000000:06:00.0 Off | 0 |
| N/A 56C P0 117W / 149W | 3284MiB / 11441MiB | 88% Default |
+-------------------------------+----------------------+----------------------+
| 1 Tesla K80 Off | 00000000:07:00.0 Off | 0 |
| N/A 36C P0 73W / 149W | 764MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 2 Tesla K80 Off | 00000000:0A:00.0 Off | 0 |
| N/A 44C P0 56W / 149W | 764MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 3 Tesla K80 Off | 00000000:0B:00.0 Off | 0 |
| N/A 40C P0 72W / 149W | 764MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 4 Tesla K80 Off | 00000000:10:00.0 Off | 0 |
| N/A 43C P0 55W / 149W | 764MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 5 Tesla K80 Off | 00000000:11:00.0 Off | 0 |
| N/A 36C P0 71W / 149W | 764MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 6 Tesla K80 Off | 00000000:86:00.0 Off | 0 |
| N/A 48C P0 57W / 149W | 764MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 7 Tesla K80 Off | 00000000:87:00.0 Off | 0 |
| N/A 41C P0 74W / 149W | 762MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 8 Tesla K80 Off | 00000000:8A:00.0 Off | 0 |
| N/A 48C P0 56W / 149W | 764MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 9 Tesla K80 Off | 00000000:8B:00.0 Off | 0 |
| N/A 39C P0 70W / 149W | 762MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 10 Tesla K80 Off | 00000000:91:00.0 Off | 0 |
| N/A 50C P0 52W / 149W | 764MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
| 11 Tesla K80 Off | 00000000:92:00.0 Off | 0 |
| N/A 39C P0 69W / 149W | 762MiB / 11441MiB | 0% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 0 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 270MiB |
| 1 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 1 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 2 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 3 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 4 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 5 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 6 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 7 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 8 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 9 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 10 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36418 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36419 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36420 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36421 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36422 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36423 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36424 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36425 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36426 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36427 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36428 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
| 11 36429 C /home/chenwei/software/deepmd-kit/bin/lmp 60MiB |
+-----------------------------------------------------------------------------+
It seems that most of the CPU time is spent on communication. How could I reduce it?
Best wishes,
Wei
Beta Was this translation helpful? Give feedback.
All reactions