cpp-performance-lab

A C++ microbenchmark repository for cache behavior, memory access, synchronization, communication paths, language/runtime overhead, allocator tradeoffs, container lookup, and syscall or network boundary cost.

Start Here

• Results summary: primary document for benchmark findings, plots, code links, and conclusions
• README.md: build, run, and repository navigation

Purpose

• Build intuition for cache hierarchy and memory access patterns
• Compare concurrency, communication, language, container, and allocator tradeoffs with reproducible microbenchmarks
• Measure syscall, IPC, and local transport overhead with small focused benchmarks
• Produce evidence-based performance notes from stable runs

1) Prerequisites

• CMake >= 3.20
• A C++20 compiler (clang++ or g++)
• Git + internet access (for fetching google/benchmark)

2) Build

cmake -S . -B build -DCMAKE_BUILD_TYPE=Release
cmake --build build -j

3) Run all benchmarks (recommended)

scripts/run_all.sh

4) Run a specific benchmark

Standard pattern:

./build/benchmark/<binary_name> --benchmark_min_time=0.3s

Examples:

./build/benchmark/bm_stride_access --benchmark_min_time=0.3s
./build/benchmark/bm_cache_levels --benchmark_min_time=0.3s

Queue tuned run:

./build/benchmark/bm_queue \
  --benchmark_filter='BM_Queue(MutexTransfer/batch:64/backoff:0|SpscRingTransfer/batch:8/backoff:0)$' \
  --benchmark_min_time=1s \
  --benchmark_repetitions=10 \
  --benchmark_report_aggregates_only=true

5) Results

• Results summary: grouped benchmark findings, plots, jump index, and direct source links

Generate figures from benchmark runs:

python3 scripts/generate_plots.py

Generate mechanism diagrams used by the summary:

python3 scripts/generate_mechanism_diagrams.py

6) Benchmark Areas

• benchmark/cache/: cache, locality, and working-set behavior
• benchmark/layout/: data layout experiments
• benchmark/concurrency/: synchronization, queues, and thread placement
• benchmark/memory/: allocator and pooling behavior
• benchmark/cpu/: instruction-throughput experiments
• benchmark/containers/: container and lookup tradeoff benchmarks
• benchmark/language/: dispatch and callable abstraction benchmarks
• benchmark/object_model/: return-value optimization and object model behavior
• benchmark/syscalls/: file and syscall boundary measurements
• benchmark/ipc/: communication-path benchmarks
• benchmark/network/: local socket and transport-path benchmarks

7) Figures

Cache Levels

Cache Associativity

Queue and Memory Pool