Profiling with perf and valgrind — reading the output
A practical guide to perf and valgrind: what each tool measures, which commands to run, and how to read the output to find real performance bottlenecks.
// categories
Performance
Cache-friendly data structures: AoS vs SoA in practice
How CPU cache lines determine whether your data layout helps or hurts performance. Array of Structures vs Structure of Arrays — with measurements.
Memory allocators: pool, arena, and avoiding heap fragmentation
Why the default heap breaks embedded firmware and high-throughput servers. Pool allocators, arena allocators, and stack allocators — implementation, tradeoffs, and when to use each.
Move semantics and zero-copy data transfer
How move semantics eliminate copies in C++11 and beyond. Rvalue references, move constructors, std::move, and the patterns that make hot paths allocation-free.
Lock-free queues — SPSC, MPSC, and when to use each
How single-producer/single-consumer and multi-producer queues work, the memory ordering guarantees they need, and the performance implications on modern CPUs.