Devoxx Greece 2024 - Busy Architect's Guide to Distributed Systems by Ted Neward

A distributed system is any program that can be run on more than one computer.

There are various technical fallacies of distributed computing, including “nothing but a bunch of networking”.

The concept of durable state is important in distributed systems, as it is the state that we really care about.

We need to think about transient state, which is the state that can change quickly.

RESTful APIs are still widely used but have limitations.

JSON is a simple, elegant data format but has limitations.

There are various architectural styles that can be used in distributed systems, including client-server and peer-to-peer.

The network can fail, and we need to be prepared for this.

Distributed systems can be single-hop (e.g., a single network) or multi-hop (e.g., multiple networks).

The shape of the state in a distributed system can be difficult to understand and manage.

We need to think about how to manage state in a distributed system, including using buffering, caching, and other techniques.

There are various ways to build a distributed system, each with its own trade-offs.

The 12-factor app is an example of a successful distributed system.

The concept of “black box” and “white box” testing can be useful in testing distributed systems.

There are various ways to manage errors and exceptions in a distributed system.

The importance of testing in a distributed system.