demonstrates that adding up certain values becomes much faster when an array is sorted, due to modern CPU branch prediction (see https://stackoverflow.com/questions/11227809/why-is-it-faster-to-process-a-sorted-array-than-an-unsorted-array)

branchprediction_problem.jl

function genlist()
    arr = zeros(32768)
    for i=1:length(arr)
        arr[i] = rand(Int) % 256
    end
    arr
end

function processlist(arr)
    sum = 0

    for i=1:length(arr)
        if arr[i] >= 128
            sum += arr[i]
        end
    end

    sum
end

x1 = genlist()
x2 = sort(x1)
# julia> using BenchmarkTools
# julia> @btime processlist($x1)
  #124.064 μs (0 allocations: 0 bytes)
#1.55527e6

#julia> @btime processlist($x2)
  #65.628 μs (0 allocations: 0 bytes)
#1.55527e6