HIGH-PERFORMANCE LINUX SERVERS: Tune CPU, Memory, Storage, Networking, Filesystems, Kernel Parameters, Containers, Databases, and Web Workloads for Maximum Performance - Tapa blanda

Sinopsis

HIGH-PERFORMANCE LINUX SERVERS is a practical, production-focused guide to diagnosing, tuning, and validating Linux server performance across the full stack.

Rather than relying on generic tuning advice, this book teaches a repeatable workflow: establish a baseline, identify the real bottleneck, make a controlled change, rebenchmark, and validate the result.

You will learn how to optimize CPU scheduling, memory pressure, NUMA locality, NVMe storage, filesystems, TCP networking, NIC queues, containers, web servers, databases, and application workloads using tools such as perf, PSI, fio, iperf3, bpftrace, BCC, bpftool, stress-ng, Prometheus, and Grafana.

The book also covers Docker and Podman resource controls, NGINX and HAProxy performance, PostgreSQL, MySQL/MariaDB, Redis, cgroup v2, systemd limits, eBPF, XDP, io_uring, IRQ placement, and workload-specific kernel tuning.

Hands-on labs show you how to reproduce and diagnose CPU saturation, memory pressure, cgroup throttling, storage bottlenecks, packet loss, socket queue problems, database latency, and other real performance issues.

The final capstone brings everything together in a complete zero-to-high-performance Linux server build, where you will baseline, break, diagnose, tune, stress-test, and validate an entire production-style stack.

Designed for Linux administrators, DevOps engineers, SREs, infrastructure engineers, platform engineers, and advanced Linux users, this book helps you move beyond guesswork and build servers whose performance can be measured, explained, and proven.

"Sinopsis" puede pertenecer a otra edición de este libro.