on ir71.eecs with 0.7.0:
CPU: Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz

@nbinclude("sum-simd.ipynb")
f = sum0
  0.168123 seconds (500 allocations: 7.813 KiB)
f = sum_in_simd
  0.158031 seconds (500 allocations: 7.813 KiB)
f = sum_in
  0.538791 seconds (500 allocations: 7.813 KiB)
f = sum_range_simd
  0.538910 seconds (500 allocations: 7.813 KiB)
f = sum_range_inbounds
  0.538775 seconds (500 allocations: 7.813 KiB)
f = sum_range_inbounds_simd
  0.157895 seconds (500 allocations: 7.813 KiB)


using sum_simd.m with 2017a :

@sum0 0.06 sec
@sum_range 2.3 sec