This shows you the differences between two versions of the page.
Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
institute_lorentz:xmaris [2022/06/10 08:40] – [Web access] lenocil | institute_lorentz:xmaris [2024/02/29 14:16] (current) – [Xmaris Partitions] jansen | ||
---|---|---|---|
Line 18: | Line 18: | ||
Xmaris is the successor of the maris cluster, renamed with a prefix '' | Xmaris is the successor of the maris cluster, renamed with a prefix '' | ||
- | [[https:// | + | [[https:// |
===== Xmaris features and expected cluster lifetime ===== | ===== Xmaris features and expected cluster lifetime ===== | ||
Line 44: | Line 44: | ||
^Mount Point^ Type ^Notes^ | ^Mount Point^ Type ^Notes^ | ||
|/scratch | HD | **temporary**, | |/scratch | HD | **temporary**, | ||
- | |/marisdata |NetApp| 2TB/user quota, remote| | + | |/marisdata |NetApp| 2TB/user quota, medium-term storage, remote| |
- | |/home |NetApp| 10GB/user quota, remote| | + | |/home |NetApp| 10GB/user quota, medium-term storage, remote| |
+ | |/ | ||
Extra efficient scratch spaces are available to all nodes on the infiniband network ('' | Extra efficient scratch spaces are available to all nodes on the infiniband network ('' | ||
^Mount Point^ Type^ Notes^ | ^Mount Point^ Type^ Notes^ | ||
- | |/IBSSD| SSD |**temporary**, InfiniBand/ | + | |/IBSSD| SSD |**DISCONTINUED**, InfiniBand/ |
|/PIBSSD| SSD|**temporary**, | |/PIBSSD| SSD|**temporary**, | ||
Line 58: | Line 60: | ||
xmaris users are strongly advised they delete (or at least move to the shared data disk), if any, their data from the compute nodes scratch disks upon completion of their calculations. All data on the scratch disks __might be cancelled without prior notice__. | xmaris users are strongly advised they delete (or at least move to the shared data disk), if any, their data from the compute nodes scratch disks upon completion of their calculations. All data on the scratch disks __might be cancelled without prior notice__. | ||
- | Note that **disk policies might change at any time at the discretion of the cluster owners**. | + | Note that **disk policies might change at any time at the discretion of the cluster owners**. |
Line 115: | Line 117: | ||
Xmaris services, that is terminal, scheduler/ | Xmaris services, that is terminal, scheduler/ | ||
- | {{ : | + | {{ : |
- | Similarly to a traditional shell access, Xmaris OpenOnDemand is available only for connections within the __IL intranet__. IL users who wish to access OpenOnDemand from their remote home locations could for example instruct their browsers to SOCKS-proxy their connections via our SSH server. | + | Similarly to a traditional shell access, Xmaris OpenOnDemand is available only for connections within the __IL intranet__. IL users who wish to access OpenOnDemand from their remote home locations could for example |
- | Open a local terminal and type | + | Open a local terminal and type (substitute username with your IL username) |
<code bash> | <code bash> | ||
Line 132: | Line 134: | ||
* Submit batch jobs to the slurm scheduler/ | * Submit batch jobs to the slurm scheduler/ | ||
* Open a terminal. | * Open a terminal. | ||
- | * Launch interactive jupyter notebooks. | + | * Launch interactive |
* Monitor cluster usage. | * Monitor cluster usage. | ||
* Create and launch your very own OnDemand application (read [[https:// | * Create and launch your very own OnDemand application (read [[https:// | ||
Line 140: | Line 142: | ||
^Partition^ Number nodes ^ Timelimit ^ Notes^ | ^Partition^ Number nodes ^ Timelimit ^ Notes^ | ||
- | |compAMD*| | + | |compAMD*| |
|compAMDlong| 3 | 60 days | | | |compAMDlong| 3 | 60 days | | | ||
|compIntel| 2 | 5 days and 12 hours| | | |compIntel| 2 | 5 days and 12 hours| | | ||
|gpuIntel| 1 | 3 days and 12 hours | GPU | | |gpuIntel| 1 | 3 days and 12 hours | GPU | | ||
- | |ibIntel | 4 | 7 days | InfiniBand, Multiprocessing | | + | |ibIntel | 8 | 7 days | InfiniBand, Multiprocessing | |
*: default partition | *: default partition | ||
Line 153: | Line 155: | ||
|maris075 |gpuIntel|2 x Nvidia Tesla P100 16GB | 6.0| | |maris075 |gpuIntel|2 x Nvidia Tesla P100 16GB | 6.0| | ||
- | Use the '' | + | Xmaris GPUs must be allocated using slurm' |
+ | <code bash> | ||
+ | srun -p gpuIntel --gres=gpu: | ||
+ | </ | ||
===== Xmaris scientific software ===== | ===== Xmaris scientific software ===== | ||
Line 190: | Line 195: | ||
Any pre-installed software can be made available in your environment via the '' | Any pre-installed software can be made available in your environment via the '' | ||
- | It is possible to save a list of modules you use often in a //module collection// | + | It is possible to save a list of modules you use often in a //module collection// |
<code bash> | <code bash> | ||
module load mod1 mod2 mod3 mod4 mod5 mod6 | module load mod1 mod2 mod3 mod4 mod5 mod6 | ||
Line 206: | Line 211: | ||
|TensorFlow-1.15.0-Miniconda3/ | |TensorFlow-1.15.0-Miniconda3/ | ||
- | To create | + | The following example shows how you can create a tensorflow-aware jupyter |
<code bash> | <code bash> | ||
- | # only on maris075 (GPU node) | + | # We use maris075 (GPU node) and load the optimised tf module |
ml load TensorFlow/ | ml load TensorFlow/ | ||
- | pip install --user | + | |
- | pip install --user jupyter-client==5.3.1 | + | # We install ipykernel, because necessary to run py notebooks |
- | ipython kernel install | + | python -m pip install |
+ | |||
+ | # We create a kernel called TFQuantum based on python from TensorFlow/ | ||
+ | python -m ipykernel | ||
+ | |||
+ | # We edit the kernel such that it does not execute python directly | ||
+ | # but via a custom wrapper script | ||
+ | cat $HOME/ | ||
+ | |||
+ | { | ||
+ | " | ||
+ | "/ | ||
+ | "-m", | ||
+ | " | ||
+ | " | ||
+ | " | ||
+ | ], | ||
+ | " | ||
+ | " | ||
+ | " | ||
+ | " | ||
+ | } | ||
+ | } | ||
+ | |||
+ | # The wrapper script will call python but only after loading any | ||
+ | # appropriate module | ||
+ | cat / | ||
+ | |||
+ | #!/bin/env bash | ||
+ | ml load TensorFlow/ | ||
+ | |||
+ | exec python $@ | ||
+ | |||
+ | # DONE. tfquantum will appear in the dropdown list of kernels | ||
+ | # upon creating a new notebook | ||
</ | </ | ||
- | When launching a jupyter notebook remember to specify '' | + | |
=== TensorFlow with Graphviz === | === TensorFlow with Graphviz === | ||
Line 235: | Line 275: | ||
* via a traditional // | * via a traditional // | ||
- | Whatever installation method you might choose, please note that you do not have administrative rights to the cluster. | + | Whatever installation method you might choose, please note that you **do not have** administrative rights to the cluster. |
Line 263: | Line 303: | ||
</ | </ | ||
- | :!: The environment variable '' | + | |:!: The environment variable '' |
- | ml|here]]. | + | ml|here]].| |
- | :!: When compiling OpenBLAS it is not sufficient to define '' | + | |:!: When compiling OpenBLAS it is not sufficient to define '' |
Then execute | Then execute | ||
Line 276: | Line 316: | ||
to make available to the '' | to make available to the '' | ||
- | :!: '' | + | |:!: '' |
Should you want to customise the building process of a given software please read how to implement [[https:// | Should you want to customise the building process of a given software please read how to implement [[https:// | ||
Line 582: | Line 622: | ||
===== Suggested readings ===== | ===== Suggested readings ===== | ||
- | * https:// | + | * https:// |
* https:// | * https:// | ||
* https:// | * https:// |