Why use a cluster? Parallel examples

[tkphd]

Right off the bat, we have a wall of text that becomes tedious to read and may not clarify how each imaginary cluster user is adapting a workflow to use a cluster. Tightening this language to more clearly describe the high-throughput, memory-intensive, and compute-intensive workloads (and perhaps mention those terms of art) would be clarifying.

[caterina159]

Hello, not sure where to post this, but it seems there is no slack platform if we get stuck, and I am trying to log into the HPC Carpentry Cloud Cluster (doing the Introduction to High-Performance Computing lesson) and I do not know where is the username given by the instructors to log on to the cluster. I have already set up my key pair.
Can you please help?

Thank you,
Caterina

[mahermanns]

Hello, not sure where to post this, but it seems there is no slack platform if we get stuck, and I am trying to log into the HPC Carpentry Cloud Cluster (doing the Introduction to High-Performance Computing lesson) and I do not know where is the username given by the instructors to log on to the cluster. I have already set up my key pair. Can you please help?

Hi @caterina159 . You can reach people better on the Carpentries Slack Workspace. There is an #hpc-carpentry channel, where you can ask for assistance with any of the material.

It seems that you are working through the material outside of a workshop setting. The HPC Carpentry material is different from other lessons in that you need access to additional resources (i.e., the HPC system) apart from your own personal computer. As it is expensive to keep the virtual cluster running all the tim, the Carpentry Cloud Cluster is only set up specifically for a workshop and inaccessible otherwise. Login credentials are then created specifically for a workshop and distributed by the workshop instructors.

We should probably clarify this in the lesson material.