Glibc's Per-Thread Cache Is Helping Out Some Benchmarks

Released this week was the shiny new glibc 2.26 GNU C Library with the notable new feature being the per-thread cache for malloc.

In most tests the glibc 2.26 upgrade isn't showing any major benefits, but for at least some there is. In particular, Redis appears to benefit greatly according to some new Phoronix Test Suite / OpenBenchmarking.org results:

In fact, going from glibc 2.25 to 2.26 now makes it faster than jemalloc.


I'll be running some tests on my end shortly, but for now there is this OpenBenchmarking.org result file.