Yeah, for compilers, I would like to see a benchmark of small string optimization (which should be a huge win for most workloads) and string interning (also sho

Yeah, for compilers, I would like to see a benchmark of small string optimization (which should be a huge win for most workloads) and string interning (also should be huge, not in the post though) A pet peeve is that benchmark suites often favor measuring performance with fibonacci or mandelbrot-type things, but string/graph/hash table workloads are much different, in multiple dimensions Compilers and network servers are more like the latter