Closes #2880: Add parallel writing when writing pdarrays to Parquet

[bmcdonald3]

This PR (closes #2880) adds support for a parallelWriteThreshold flag that
allows a user to determine the size of files of the Parquet files
to be written and then write those files in parallel, appending a
_CORE#### to the end of the file name.

By running the Arkouda server with:
./arkouda_server --ParquetMsg.parallelWriteThreshold=<num>, a
user is able to control the size of the files that are going to be
written.

This is currently only supported on pdarrays of natively-supported
datatypes (meaning not strings or dataframes), but follow work is
on the way.

[jaketrookman]

Looks good

[stress-tess]

A whole lot of comments from me all just to say... LGTM lol

Nice work ben! This looks great!! 🚀

[stress-tess]

okay this is pedantic, but this could just be var doParallel = A.size > parallelWriteThreshold; right?

[stress-tess]

i had to convince myself that the integer cast of the float divide would always round down. This seems to check out and the adjustment gave what i expected on a small example!

var fileSizes: [0..#6] int = 10/6;  // 1.6666 verify this rounds down
writeln(fileSizes);
var leftOver = 10 - ((10/6)*6); 
writeln(leftOver);

1 1 1 1 1 1
4

I will say in my small example this resulted in a pretty unbalanced distribution, but I think that in a real case that locDom.size would be large enough relative to loc.maxTaskPar that it would be pretty uniform... I'm just now realizing that's probably part of the motivation for having a parallelWriteThreshold lol

[stress-tess]

Since leftOver is the remainder of locDom.size/loc.maxTaskPar, it should be guaranteed to be less than fileSizes.size. So if we wanted to distribute the remainder more evenly we could do something like

var fileSizes: [0..#6] int = 10/6;
writeln(fileSizes);
var leftOver = 10 - ((10/6)*6); 
writeln(leftOver);
fileSizes[0..#leftOver] += 1;
writeln(fileSizes);

1 1 1 1 1 1
4
2 2 2 2 1 1

But this is probably overengineering something that isn't actually a problem

[stress-tess]

instead of doing this back by one, couldn't you do

var offsets = (+ scan fileSizes) - fileSizes;
forall (i, off, len) in zip(fileSizes.domain, offsets, fileSizes) {
...

I don't think this would have any performance difference, but this is more similar to how we calculate offsets in other places. I normally prefer looping variables over indexing when possible because it makes easier for me to tell at a glance what's local, but that doesn't apply here. So there's def no need to change

[stress-tess]

Since @bmcdonald3 is out and none of my comments are blocking, I'll go ahead and merge this. Thanks again ben!!!

EDIT: apparently ben is in and he wants to hold off for string support

[bradcray]

Reviewing PRs from our team today, I came across this one and wondered about its status. My understanding is:

• It only handles ints in its current state, while ultimately one would want to do the same for strings (though maybe that doesn't need to be done in this PR?)
• Ben handed it off to @vasslitvinov when he took his new position, though it may have fallen between the cracks with other things Vass was also handling
• Vass will be unable to work on this for the next few months, so we should consider having someone else pick it up

Tagging @e-kayrakli for awareness and @jhh67 due to his recent Parquet work (albeit in the runtime rather than module-level code).