Use GNU parallel to run consequences processing in parallel

[anthonyfok]

Taking advantage of multiple CPU cores, multiple python3 instances are
dispatched simultaneously using "GNU parallel" in run_OQStandard.sh
for consequences calculations.

Using bash scripts/run_OQStandard.sh SCM5p8_Montreal_conv -h -r -d -o
as example, with each realization taking 82 minutes, doing 16 realizations
in parallel instead of in series would save 20.5 hours. As consequences
calculations are done twice, the total run time is reduced by 41 hours,
from 56 hours down to 15 hours on a c5a.24xlarge EC2 instance.

Unlike Python’s own multiprocessing module, GNU parallel’s invocation of
multiple invocations of Python does not involve any memory sharing at all,
which avoids any potential mysterious calculation discrepancy with
Numpy’s OpenBLAS dot multiplications seen in superseded Pull Request #58.

Fixes #57

---

Hey @tieganh and @jeremyrimando,

• For convenience, this PR builds upon and includes the proposed changes in PR run_OQStandard.sh: Restore "oq engine --run" for Hazard calculation #56
• This code is being used by @jeremyrimando for scenario calculations since Monday, May 9.
• The code is far from perfect, but I thought I'd better not create unneeded delays for @jeremyrimando, and in order to help match the actual running code with that in the Git repo, I decided to commit the code as-is.

All comments welcome!

[jvanulde]

@anthonyfok any value in sharing these findings with @micheles at GEM?

[anthonyfok]

@anthonyfok any value in sharing these findings with @micheles at GEM?

@jvanulde Great idea! Sorry for taking so long to get back to you. I'll try to send an email to Micheles et al. today

---

2024-01-09 update: Didn't email Micheles directly, but finally opened this following issue:

• Some notes on parallelizing Consequences calculations gem/oq-engine#9324