anomymous workshop feedback, #3 #11

psteinb · Mar 20, 2017

excerpt from a evaluation feedback regarding the first run of this workshop in March 2017

The introduction of the shell was not necessary. That should be a prerequisite. As a consequence, the actual parallel programming went way too fast. The worked example (estimation of Pi) is a case where one computation is split and then the results pulled together for the final result. It would have been nice to have an example where one task is distributed onto many nodes and output is pulled together. Furthermore, MPI needs more in depth treatment and Snakemake was mentioned but not presented. That looks like a nice library too. In summary, too little time was devoted to HPC and too much to shell operations.

psteinb changed the title from anomymous workshop feedback, #2 to anomymous workshop feedback, #3 Mar 20, 2017

psteinb/hpc-in-a-day

anomymous workshop feedback, #3 #11

psteinb commented Mar 20, 2017

psteinb changed the title from anomymous workshop feedback, #2 to anomymous workshop feedback, #3 Mar 20, 2017

psteinb/hpc-in-a-day

Join GitHub today

anomymous workshop feedback, #3 #11

Comments

psteinb commented Mar 20, 2017

psteinb changed the title from anomymous workshop feedback, #2 to anomymous workshop feedback, #3 Mar 20, 2017