Follow rule for what to put in `cluster_guides` and what in `software` folder

[richelbilderbeek]

Currently, it is unclear which files should be where.

For example, this is the current cluster_guides folder:

richel@richel-N141CU:~/GitHubs/UPPMAX-documentation/docs/cluster_guides$ ls
bianca_file_transfer_using_gui.md        running_jobs
bianca.md                                screen.md
bianca_modules.md                        singularity.md
bianca_overview.md                       slurm.md
biancas_design.md                        slurm_on_bianca.md
biancas_name.md                          slurm_on_rackham.md
compilers.md                             slurm_on_snowy.md
compiling_parallel.md                    slurm_troubleshooting.md
compiling_serial.md                      snowy.md
conda.md                                 snowys_design.md
dardel.md                                snowys_name.md
dardel_migration.md                      start_interactive_node.md
darsync.md                               start_interactive_node_on_bianca.md
debuggers.md                             start_interactive_node_on_rackham.md
empty.md                                 start_interactive_node_on_snowy.md
ides_on_bianca.md                        thinlinc.md
ides_on_rackham.md                       transfer_bianca.md
img                                      transfer_dardel.md
install.md                               transfer_rackham.md
login_node_restrictions.md               transfer_transit.md
modules.md                               transit_file_transfer_using_gui.md
old                                      transit_file_transfer_using_scp.md
optimizing_jobs.md                       transit_file_transfer_using_sftp.md
profilers.md                             transit.md
project_management.md                    uppmax_as_an_organization.md
rackham_file_transfer_using_gui.md       uppmax_cloud.md
rackham_file_transfer_using_scp.md       uppmax_cluster.md
rackham_file_transfer_using_sftp.md      uppmax_filesystem.md
rackham_file_transfer_using_transit.md   uppmax_history.md
rackham.md                               uppmax.md
rackham_modules.md                       uppmax_systems.md
rackham_remote_desktop_extract_file.png  venv_on_rackham.md
rackhams_design.md                       vscode_on_bianca.md
rackhams_name.md                         vscode_on_rackham.md
rstudio_on_bianca.md                     vscodium_on_bianca.md
rstudio_on_rackham.md                    vscodium_on_rackham.md

I expect to see:

• Things about specific clusters: [cluster_name].md, e.g. bianca.md
• Using software on a specific cluster: [software]_on_[clustername].md, e.g. vscodium_on_bianca.md
• General cluster things: install.md, login.md

However, these are just software:

• compilers.md
• screen.md
• singularity.md
• slurm.md

I feel these should be put on software.

That is the point of this Issue: what to put in cluster_guides and what in software?

My suggested rule:

Description	Where
Information about clusters	cluster_guides
Information about general cluster processes	cluster_guides
Cluster-specific information for software	cluster_guides
Software in general	software

[richelbilderbeek]

Björns rule:

• If the software is loaded by a module, put it in software
• Else put it in cluster_guides

[bclaremar]

Installing and dev section goes to Software, including all sub pages

[bclaremar]

slurm is kept in cluster

[richelbilderbeek]

Next week: people suggest their favorite rules and we'll vote on these :+1:

[richelbilderbeek]

Rule 1a

My suggested rule, now with some examples added.

Description	Where	Typical name
Information about clusters	cluster_guides	bianca.md, rackham.md, snowy.md
Information about general cluster processes	cluster_guides	[general_process_name].md
Cluster-specific information for software	cluster_guides	[x]_on_bianca.md
Software in general	software	[name of program].md

Example on file transfer:

• cluster_guides/bianca_file_transfer_using_rsync.md
• software/rsync.md
• cluster_guides/file_transfer.md

Example on job scheduler:

• cluster_guides/slurm_on_bianca.md
• cluster_guides/job_scheduler.md (a general cluster process)
• software/squeue.md
• software/sbatch.md
• software/slurm.md (a collection of programs is software)

Example on general software use:

• software/gcc.md (a collection of programs is software)
• software/gcc_compile_c.md (same for all clusters)

Rule 1b

Same as rule 1a, except for cluster_guides/slurm.md

[richelbilderbeek]

Rule 1c

Similar to 1a, except:

Description	Where	Typical name
Cluster-specific information for software	software	[x]_on_bianca.md

[richelbilderbeek]

Rule 1d

Combines rule 1b and 1c

[richelbilderbeek]

• Richel: 1a, 1c
• Diana: 1d
• Bjorn: 1d
• Pavlin: 1d

[richelbilderbeek]

1d it is!

[richelbilderbeek]

Richel will follow this rule and update the links!

[bclaremar]

All done and close??

[richelbilderbeek]

Still in progress, will continue after my holiday