Tag: performance

Your gigabit link drops to 200 Mbps and you don't know why. iperf3 measures throughput honestly, nload visualizes traffic, together they find the bottleneck fast.

Stop trusting marketing IOPS numbers. fio measures what your disk actually does on your workload, sequential, random, mixed. Here's how to read the output honestly.

Linux ships with conservative kernel defaults. These sysctl settings tune your server for networking, memory, and file I/O, with explanations, not just paste-and-pray values.

A kernel upgrade halved one team's PostgreSQL throughput. The fix was kernel tuning: huge pages, THP, swappiness, I/O scheduler. Here's what matters and why.

Learn multi-stage Docker builds to slash image sizes by 90%. Practical before/after examples for Node.js, Python, and Go with real size comparisons.

Learn systemd socket activation to start services on-demand, save RAM, and cut boot time. Includes .socket unit files, real examples, and testing with systemd-socket-activate.

Your CI job waits 4 minutes for npm install every run. Caching cuts it to 15 seconds. Dependency caching for GitHub Actions, GitLab CI, pip, and Docker.

zstd is as fast as gzip with near-xz compression ratios. Here's why you should drop bzip2 forever and how to use zstd in 2026.

Your Linux box has hundreds of kernel knobs in /proc/sys that nobody ever touches. Most do not matter. Here are the handful that actually improve a Docker host.

Tmpfs vs ramfs on Linux: mount RAM-backed filesystems for blazing-fast temp storage. Covers fstab, Docker tmpfs mounts, CI/CD use cases, and the key differences.

Nextcloud advanced configuration: Redis caching, federation setup, automated backups, occ command deep dive, LDAP, external storage, and PHP-FPM tuning.

ulimit and cgroups v2: set per-process CPU, memory, and file limits, use systemd slice controls, and keep one runaway service from killing your server.