MINIX 3: a Modular, Self-Healing POSIX-compatible Operating System

It's also for security. If you find a security vulnerability in e.g. the sound driver, you're not going to get root access. Andrew Tanenbaum explains it well: [VIDEO]

Andrew Tanenbaum presenting on Minix 3's architecture: [VIDEO] . Regardless of one's opinions on microkernels vs. monolithic kernels, it's a very interesting but accessible talk for those interested in lower-level systems and fault-tolerant architectures.

Here's Andy Tannenbaum's response to the performance issues of microkernels: [VIDEO] (should start at about 47:15)

Writing OS's is hard and there are no shortcuts. You should take a look at microkernels and the work Andrew Tanenbaum did with Minix 3 [1] and his modular design as this is the sort of direction you're probably looking for with customisable OS's. [1] [VIDEO]

(2015) Previously: [LINK] And see [1], my note about this talk being more or less the same as the 2010 Minix talk at FOSDEM[2]. If you saw that one, or any other talk by Tanenbaum of that period, I'll spoil it for you: there's nothing substantially different in this talk. 1. [LINK] 2. [VIDEO]

Here's you some examples to learn from. Most are old enough that patents have expired: Why Do Computers Stop and What Can Be Done About It? (1985) [LINK] Their NonStop Architecture [LINK] QNX's systems are ultra-reliable, too [LINK] OpenVMS clusters' uptime was years to decades & mixed CPU ISA's [LINK] Systems that run forever and self-heal (Armstrong) [LINK] Microreboots [LINK] Minix 3: Microkernel-based, self-healing, POSIX OS [VIDEO]