ashvardanian / HashTableBenchmark

A simple cross-platform speed & memory-efficiency benchmark for the most common hash-table implementations in the C++ world

Hash-Tables Benchmarks

This repository contains a bunch of extendable benchmarks mostly for following containers:

• std:unordered_map from STL.
• google::dense_hash_map from sparsehash/sparsehash v2.0.4.
• google::sparse_hash_map from sparsehash/sparsehash v2.0.4.
• tsl::sparse_map from tessil/sparse-map v0.6.2.
• tsl::hopscotch_map from tessil/hopscotch-map v2.3.0.
• tsl::robin_map from tessil/robin-map v0.6.3.

The keys and values are generally set to 64-bit integers to mimic most workloads. Keys are generated with the uniform ditribution. The hash function is set to std::hash.

Build

Running bash -i build.sh will build following executables:

• ./build/build/bin/mixed_ops - speed test comparing insertions, removals and lookups across containers.
• ./build/build/bin/ram_usage - a benchmark analyzing the space overhead of each container, which shouldn't be called directly, only through python ram_usage_bench.py.
• ./build/build/bin/ram_speed - a memory-bandwidth utilization benchmark.