Online / 6 & 7 February 2021


Shaken Fist, thought experiments in simpler IaaS clouds

OpenStack today is a complicated beast -- not only does it try to perform well for large clusters, but it also embraces a diverse set of possible implementations from hypervisors, storage, networking, and more. This was a deliberate tactical choice made by the OpenStack community years ago, forming a so called "Big Tent" for vendors to collaborate in to build Open Source cloud options. It made a lot of sense at the time to be honest. However, OpenStack today finds itself constrained by the large number of permutations it must support, ten years of software and backwards compatability legacy, and a decreasing investment from those same vendors that OpenStack courted so actively.

OpenStack Compute wasn't always like this though. The first public git commit had about 5,000 lines of code in it, and was surprisingly functional. What would a simpler and more opinionated IaaS compute component look like? Especially if it could take advantage of the general improvement of Open Source tooling in the last decade that OpenStack has driven? This was what I was pondering in late 2019 as I started to drift away from the OpenStack community for various reasons. Then, a series of natural disasters in Australia and globally presented me with an opportunity to cancel all my extracurricular activities and really give an experiment a go. That experiment is called Shaken Fist (as in old man shakes fist at cloud), is about the same size as the original OpenStack Compute commit, and is available at I'd appreciate the opportunity to discuss some of the simplifying assumptions made, what functionality is available, whether the experiment tells us anything interesting, and possible future directions.


Photo of Michael Still Michael Still