Summary: I describe a simple interview problem (counting frequencies of unique words), solve it in various languages, and compare performance across them. For each language, I’ve included a simple, idiomatic solution as well as a more optimized approach via profiling…

Language	Simple	Optimized	Notes
grep	0.04	0.04	grep baseline; optimized sets LC_ALL=C
wc -w	0.29	0.20	wc baseline; optimized sets LC_ALL=C
Zig	0.54		by ifreund and matu3ba
Nim	0.76	0.58	by csterritt and euantorano
C	0.97	0.23
Go	1.14	0.39
Crystal	1.29		by Andrea Manzini
PHP	1.36		by Max Semenik
Rust	1.43	0.38	by Andrew Gallant
C#	1.51	0.82	by J Taylor, Y Ostapenko, O Turan
OCaml	1.72		by Nate Dobbins and Pavlo Khrystenko
C++	1.73	0.42	optimized by Jussi Pakkanen
Perl	1.81		by Charles Randall
F#	1.82	1.59	by Yuriy Ostapenko
Kotlin	1.86		by Kazik Pogoda
Python	2.07	1.30
Lua	2.50	1.97	by themadsens; runs under luajit
JavaScript	2.52	1.90	by Dani Biro and Flo Hinze
Ruby	3.13	2.43	by Bill Mill
AWK	3.55	1.13	optimized uses mawk
D	4.16	1.01	by Ross Lonstein
Swift	4.23		by Daniel Muellenborn
Forth	4.26	1.46
Shell	14.60	1.85	optimized does LC_ALL=C sort -S 2G

This thread was posted by one of our members via one of our news source trackers.