Engineering & product insights
Deep-dives, announcements, and war stories from the team building the infrastructure.
Introducing Cloud Servers: Pay Only for What You Use
Today we're launching Cloud Servers — flexible, pay-per-hour compute that spins up in under 30 seconds. No commitments, no minimums. Here's what we built and why.
By James WhitmoreNVMe vs. SATA SSD: Why We Migrated Our Entire Storage Fleet
After benchmarking every VPS node in our fleet, we made the call to migrate everything to NVMe. The numbers were too good to ignore — and the migration took 72 hours.
Building a 1 Tbps+ DDoS Mitigation System from Scratch
We were getting hit with volumetric attacks that saturated our upstream links. This is the story of how we built a scrubbing infrastructure that can absorb over a terabit per second.
Inside Equinix FR5: Setting Up Our Frankfurt Expansion
We added 400 physical servers across three new cabinets in Frankfurt this quarter. A behind-the-scenes look at the hardware choices, network design, and BGP configuration.
How We Hit 99.99% Uptime Across 17 Regions for Two Consecutive Years
Availability at this level isn't an accident. It's the result of redundancy architecture, runbooks, chaos engineering, and a culture that treats on-call seriously.
Serversify at Five: From One Rack in Frankfurt to 17 Global Locations
Five years ago we had a single rack, two founders, and a lot of ambition. Here's an honest look at what worked, what didn't, and what we're building toward.