[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki

[ Back to the navigation ]

You are here: start » slurm

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
Next revision Both sides next revision
slurm [2022/10/17 11:41]
fucik [Interactive mode] 		slurm [2023/01/11 16:03]
krubinski [gpu-troja]
Line 5: Line 5:
 Currently there are following partitions (queues) available for computing: Currently there are following partitions (queues) available for computing:
  
-| **Partition name** | **Nodes**  | **Note** | +===== Node list by partitions ===== 
-| cpu-troja      | 7x CPU | default partition | + 
-| gpu-troja      | 6x GPU | features: gpuram48G,gpuram40G | +The naming convention is straightforward for CPU nodes - nodes in each group are numbered. For GPU nodes the format is: [t]dll-**X**gpu**N** where **X** gives total number of GPUs equipped and **N** is just enumerating the order of the node with the given configuration. 
-| gpu-ms         | 7x GPU | features: gpuram48G,gpuram24G |+The prefix **t** is for nodes at Troja and **dll** stands for Deep Learning Laboratory.  
 +==== cpu-troja ==== 
 + 
 +| Node name | Thread count | Socket:Core:Thread | RAM (MB) | 
 +| achilles1 | 32 | 2:8:2 | 128810 | 
 +| achilles2 | 32 | 2:8:2 | 128810 | 
 +| achilles3 | 32 | 2:8:2 | 128810 | 
 +| achilles4 | 32 | 2:8:2 | 128810 | 
 +| achilles5 | 32 | 2:8:2 | 128810 | 
 +| achilles6 | 32 | 2:8:2 | 128810 | 
 +| achilles7 | 32 | 2:8:2 | 128810 | 
 +| achilles8 | 32 | 2:8:2 | 128810 | 
 +| hector1 | 32 | 2:8:2 | 128810 | 
 +| hector2 | 32 | 2:8:2 | 128810 | 
 +| hector3 | 32 | 2:8:2 | 128810 | 
 +| hector4 | 32 | 2:8:2 | 128810 | 
 +| hector5 | 32 | 2:8:2 | 128810 | 
 +| hector6 | 32 | 2:8:2 | 128810 | 
 +| hector7 | 32 | 2:8:2 | 128810 | 
 +| hector8 | 32 | 2:8:2 | 128810 | 
 +| helena1 | 32 | 2:8:2 | 128811 | 
 +| helena2 | 32 | 2:8:2 | 128811 | 
 +| helena3 | 32 | 2:8:2 | 128811 | 
 +| helena4 | 32 | 2:8:2 | 128811 | 
 +| helena5 | 32 | 2:8:2 | 128810 | 
 +| helena6 | 32 | 2:8:2 | 128811 | 
 +| helena7 | 32 | 2:8:2 | 128810 | 
 +| helena8 | 32 | 2:8:2 | 128811 | 
 +| paris1 | 32 | 2:8:2 | 128810 | 
 +| paris2 | 32 | 2:8:2 | 128810 | 
 +| paris3 | 32 | 2:8:2 | 128810 | 
 +| paris4 | 32 | 2:8:2 | 128810 | 
 +| paris5 | 32 | 2:8:2 | 128810 | 
 +| paris6 | 32 | 2:8:2 | 128810 | 
 +| paris7 | 32 | 2:8:2 | 128810 | 
 +| paris8 | 32 | 2:8:2 | 128810 | 
 +| hyperion2 | 64 | 2:16:2 | 257667 | 
 +| hyperion3 | 64 | 2:16:2 | 257667 | 
 +| hyperion4 | 64 | 2:16:2 | 257667 | 
 +| hyperion5 | 64 | 2:16:2 | 257667 | 
 +| hyperion6 | 64 | 2:16:2 | 257667 | 
 +| hyperion7 | 64 | 2:16:2 | 257667 | 
 +| hyperion8 | 64 | 2:16:2 | 257667 | 
 +==== cpu-ms ==== 
 + 
 +| Node name | Thread count | Socket:Core:Thread | RAM (MB) | 
 +| iridium | 16 | 2:4:2 | 515977 | 
 +| orion1 | 40 | 2:10:2 | 128799 | 
 +| orion2 | 40 | 2:10:2 | 128799 | 
 +| orion3 | 40 | 2:10:2 | 128799 | 
 +| orion4 | 40 | 2:10:2 | 128799 | 
 +| orion5 | 40 | 2:10:2 | 128799 | 
 +| orion6 | 40 | 2:10:2 | 128799 | 
 +| orion7 | 40 | 2:10:2 | 128799 | 
 +| orion8 | 40 | 2:10:2 | 128799 | 
 +==== gpu-troja ==== 
 + 
 +| Node name | Thread count | Socket:Core:Thread | RAM (MB) | Features | GPU type | 
 +| tdll-3gpu1 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | NVIDIA A40 | 
 +| tdll-3gpu2 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | NVIDIA A40 | 
 +| tdll-3gpu3 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | NVIDIA A40 | 
 +| tdll-3gpu4 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | NVIDIA A40 | 
 +| tdll-8gpu1 | 64 | 2:16:2 | 257666 | gpuram40G gpu_cc8.0 | NVIDIA A100 | 
 +| tdll-8gpu2 | 64 | 2:16:2 | 257666 | gpuram40G gpu_cc8.0 | NVIDIA A100 | 
 +| tdll-8gpu3 | 32 | 2:8:2 | 253725 | gpuram16G gpu_cc7.5 | NVIDIA Quadro P5000 | 
 +| tdll-8gpu4 | 32 | 2:8:2 | 253725 | gpuram16G gpu_cc7.5 | NVIDIA Quadro P5000 | 
 +| tdll-8gpu5 | 32 | 2:8:2 | 253725 | gpuram16G gpu_cc7.5 | NVIDIA Quadro P5000 | 
 +| tdll-8gpu6 | 32 | 2:8:2 | 253725 | gpuram16G gpu_cc7.5 | NVIDIA Quadro P5000 | 
 +| tdll-8gpu7 | 32 | 2:8:2 | 253725 | gpuram16G gpu_cc7.5 | NVIDIA Quadro P5000 | 
 +==== gpu-ms ==== 
 + 
 +| Node name | Thread count | Socket:Core:Thread | RAM (MB) | Features | 
 +| dll-3gpu1 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | 
 +| dll-3gpu2 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | 
 +| dll-3gpu3 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | 
 +| dll-3gpu4 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | 
 +| dll-3gpu5 | 64 | 2:16:2 | 128642 | gpuram48G gpu_cc8.6 | 
 +| dll-4gpu1 | 40 | 2:10:2 | 187978 | gpuram24G gpu_cc8.6 | 
 +| dll-4gpu2 | 40 | 2:10:2 | 187978 | gpuram24G gpu_cc8.6 | 
 +| dll-8gpu1 | 64 | 2:16:2 | 515838 | gpuram24G gpu_cc8.0 | 
 +| dll-8gpu2 | 64 | 2:16:2 | 515838 | gpuram24G gpu_cc8.0 | 
 +| dll-8gpu3 | 32 | 2:8:2 | 257830 | gpuram16G gpu_cc8.6 | 
 +| dll-8gpu4 | 32 | 2:8:2 | 253721 | gpuram16G gpu_cc8.6 | 
 +| dll-8gpu5 | 40 | 2:10:2 | 385595 | gpuram16G gpu_cc7.5 | 
 +| dll-8gpu6 | 40 | 2:10:2 | 385595 | gpuram16G gpu_cc7.5 | 
 +| dll-10gpu1 | 32 | 2:8:2 | 257830 | gpuram16G gpu_cc8.6 | 
 +| dll-10gpu2 | 32 | 2:8:2 | 257830 | gpuram11G gpu_cc6.1 | 
 +| dll-10gpu3 | 32 | 2:8:2 | 257830 | gpuram11G gpu_cc6.1 | 
 + 
 + 
 +==== Submit nodes ==== 
  
 In order to submit a job you need to login to one of the head nodes: In order to submit a job you need to login to one of the head nodes:
Line 14: Line 105:
    lrc1.ufal.hide.ms.mff.cuni.cz    lrc1.ufal.hide.ms.mff.cuni.cz
    lrc2.ufal.hide.ms.mff.cuni.cz    lrc2.ufal.hide.ms.mff.cuni.cz
 +   sol1.ufal.hide.ms.mff.cuni.cz
 +   sol2.ufal.hide.ms.mff.cuni.cz
 +   sol3.ufal.hide.ms.mff.cuni.cz
 +   sol4.ufal.hide.ms.mff.cuni.cz
 ===== Basic usage ===== ===== Basic usage =====
  
Line 157: Line 252:
 There are many more parameters available to use. For example: There are many more parameters available to use. For example:
  
-<code>srun -p cpu-troja --mem=64G --pty bash</code>+**To get an interactive CPU job with 64GB of reserved memory:** 
 +<code>srun -p cpu-troja,cpu-ms --mem=64G --pty bash</code>
  
   * ''-p cpu-troja'' explicitly requires partition ''cpu-troja''. If not specified slurm will use default partition.   * ''-p cpu-troja'' explicitly requires partition ''cpu-troja''. If not specified slurm will use default partition.
   * ''--mem=64G'' requires 64G of memory for the job   * ''--mem=64G'' requires 64G of memory for the job
  
-To get interactive job with a single GPU of any kind:+**To get interactive job with a single GPU of any kind:**
 <code>srun -p gpu-troja,gpu-ms --gres=gpu:1 --pty bash</code> <code>srun -p gpu-troja,gpu-ms --gres=gpu:1 --pty bash</code>
   * ''-p gpu-troja,gpu-ms'' require only nodes from these two partitions   * ''-p gpu-troja,gpu-ms'' require only nodes from these two partitions
Line 174: Line 270:
 <code>srun -p gpu-troja --constraint="gpuram48G|gpuram40G" --mem=64G --gres=gpu:2 --pty bash</code> <code>srun -p gpu-troja --constraint="gpuram48G|gpuram40G" --mem=64G --gres=gpu:2 --pty bash</code>
   * ''--constraint="gpuram48G|gpuram40G"'' only consider nodes that have either ''gpuram48G'' or ''gpuram40G'' feature defined   * ''--constraint="gpuram48G|gpuram40G"'' only consider nodes that have either ''gpuram48G'' or ''gpuram40G'' feature defined
 +
 +==== Delete Job ====
 +<code>scancel <job_id> </code>
  
 To see all the available options type: To see all the available options type:
  
 <code>man srun</code> <code>man srun</code>
 +
 +===== See also =====
 +
 +https://www.msi.umn.edu/slurm/pbs-conversion
 +
  

[ Back to the navigation ] [ Back to the content ]