Update parallel-computing.md

[OkonSamuel]

I got a different result on my system when i ran the following codeblock

julia> i = Threads.Atomic{Int}(0);

julia> ids = zeros(4);

julia> old_is = zeros(4);

julia> Threads.@threads for id in 1:4
           old_is[id] = Threads.atomic_add!(i, id)
           ids[id] = id
       end

The result gotten in the julia manual is

julia> old_is
4-element Array{Float64,1}:
 0.0
 1.0
 7.0
 3.0

While on my system i got

julia> ids
4-element Array{Float64,1}:
 0.0
 1.0
 3.0
 6.0

[Keno]

Either result is fine, the output depends on the thread execution order. Of course that makes it a bit of a confusing example in this place. @kshyatt @vchuravy Looks like you wrote this section. What was the intention behind this example.

[vchuravy]

Maybe we should remove that example and use the one afterwards since it is independent of order of execution. I believe that particular one is from @kshyatt so let's wait for her to chime in.

[OkonSamuel]

Ok

[ViralBShah]

How about just mention for this example that the results will be printed in order of execution, and that it may different every time the program is run?

[vtjnash]

I don't think we need any updates, since the point is to make people avoid this unless they understand race conditions. Although I can include changes in #37847 (where I'm already editing this) if something is desired.