diff options
Diffstat (limited to 'README.md')
| -rw-r--r-- | README.md | 5 |
1 files changed, 4 insertions, 1 deletions
@@ -115,6 +115,8 @@ int main(void) { | Ada Hashed_Maps (GNAT) | ~147 | [4] | | std::unordered_map | ~176 | | | Python dict | ~193 | | +| zsh assoc array | ~5100 | [5] | +| bash assoc array | ~7000 | [5] | > **Note on extended L3 cache:** The test machine has a split cache topology where the benchmark's ~32 MB working set fits comfortably in one cache domain but thrashes in the other. Best times above reflect lucky scheduling onto the larger cache. Average times are roughly 1.5-2x higher. Pin with `numactl --cpunodebind=0` for reproducible results. @@ -123,7 +125,8 @@ int main(void) { **[1]** abseil is the reference Swiss table implementation. **[2]** std::unordered_map is chained; mimalloc recovers ~38% by reducing per-node allocation cost. Without it: ~176 ms. **[3]** D's built-in associative array uses per-entry heap allocation; no open-addressing or SIMD. -**[4]** Ada's `Ada.Containers.Hashed_Maps` is also chained; `Hash_Type` is 32-bit so the mixing hash is truncated, reducing distribution quality slightly. +**[4]** Ada's `Ada.Containers.Hashed_Maps` is also chained; `Hash_Type` is 32-bit so the mixing hash is truncated, reducing distribution quality slightly. +**[5]** Measured at N=10,000 (100x smaller) and extrapolated. Shells are not meant for this. --- |