Benchmark effects of parallelism on status checking

[vyasr]

By default, we perform thread-parallel execution of status fetching operations, but it's not clear that this mode is always desirable relative to serial execution since it only accelerates highly I/O bound tasks (in Python) and introduces significant overhead. We should benchmark this to test.

Original comment

Good question. I aimed to make my benchmarks representative-ish, but I would like one or two real-world benchmarks from users before making that decision. Ideally something on an HPC cluster with a "realistic" file system (not my local M.2 SSD). Would you be willing to try both options on a project of yours?

Originally posted by @bdice in #417 (comment)