Reviewed a paper analyzing lots of HPC system telemetry. Authors developed new ways to solve challenges in the data. It’s well written and reflects diligence and care. But the authors never explain why any of the results actually matter to #HPC users/operators/designers. It’s one of those things where you say “that’s neat!” and move on, because the findings don’t actually have clear impact.

Paper got rejected after other reviewers read my review, but I’ve got heartburn.