Test-Drive Performance
summary
Concrete microbenchmarks compare vector vs deque as a stack for depth‑first traversal when child counts are unknown, using the LVM‑derived example code on GitHub, seeding the PRNG, timing seconds over millions of iterations. Results show deque wins when frequent reallocations occur (about a 30% improvement in one test), but adding reserve changes performance (vector becomes up to ~40% faster in another run); tests use PODs and note movable vs copyable types matter.
AI-written summary · about
chapters (5)
related talks
01h 00mMoving Faster: Everyday efficiency in modern C++
59mPerformance matters - new tricks for old dogs
01h 01mRefresher on Containers, Algorithms and Performance in C++
01h 00mCache Friendly C++
01h 01mFaster, Easier, Simpler Vectors
01h 04mDesign for Performance
18mDistVS: Large-scale Vector Search with Compute-Memory Disaggregation
01h 29mWhen Nanoseconds Matter: Ultrafast Trading Systems in C++