Operating System Internals & Linux Kernel Bypass
“Systems Performance: Enterprise and the Cloud” by Brendan Gregg
Section titled ““Systems Performance: Enterprise and the Cloud” by Brendan Gregg”Written by the world’s leading expert on performance profiling, this book covers OS internals thoroughly. To build ultra-low latency systems, you must know how the Linux kernel schedules threads, manages virtual memory paging, and handles context switches. This book teaches you how to look for bottlenecks using eBPF, perf, and flame graphs.
Erik Rigtorp’s Engineering Blog & Repositories
Section titled “Erik Rigtorp’s Engineering Blog & Repositories”The personal technical blog and open-source library collection of low-latency systems engineer Erik Rigtorp. Rigtorp provides incredibly clean, production-ready reference code for components critical to HFT, such as ultra-fast Single-Producer Single-Consumer (SPSC) queues, lock-free ring buffers, and custom hash maps.
Network Stack Bypass Frameworks (Solarflare ef_vi / DPDK)
Section titled “Network Stack Bypass Frameworks (Solarflare ef_vi / DPDK)”Documentation and development kits for hardware-level networking. Standard Linux TCP/IP stacks introduce milliseconds of delay. Top-tier systems utilize Kernel Bypass via Solarflare’s ef_vi or Intel’s DPDK to pull network packets directly from the Network Interface Card (NIC) into your C++ application space.