Afterword

We began this journey with a simple observation: the Internet has changed what it means to use a computer. Over the course of forty-two chapters, we have built — piece by piece — the machinery that makes planetary scale computing possible. From serialization formats and RPC protocols to consensus algorithms, from caching layers and storage engines to the physical infrastructure of servers, buildings, and power systems.

If there is one lesson I hope you take away, it is that these systems are not magic. They are the work of engineers making deliberate trade-offs under constraints. Every system we studied is a choice: consistency or availability, latency or throughput, simplicity or flexibility. The art of systems engineering is not in finding the perfect answer but in understanding which trade-off is right for the problem at hand.

The field is still young. We are building the first generation of truly planetary scale computers, and many of the hardest problems remain open. How do we build systems that span jurisdictions with different privacy laws? How do we make planetary scale computing sustainable in its energy consumption? How do we ensure that these powerful systems serve all of humanity, not just those who can afford to build them?

I wrote this book because I believe that understanding how these systems work should not be a privilege reserved for engineers at the largest companies. The techniques are not secret — they are earned through experience, shared through papers and postmortems, and refined through years of operation. My hope is that this book has made some of that knowledge more accessible, and that you will use it to build systems that are not only reliable and efficient but also humane.

The planetary scale computer is not a destination. It is an ongoing project, built and rebuilt by each generation of engineers who inherit it. I am glad to have shared a small part of that project with you.

— Justin J. Meza, San Francisco, 2025