Mesos and other resource scheduling technologies will be an essential part of the next generation data center platforms.
In my previous post, I provided the proverbial 10,000 foot overview of Apache Mesos and it’s value as a resource manager for the data center. It’s time to dig a little deeper into the technology and walk though the components that make Mesos such a great candidate for a data center OS kernel. Most of the details I am walking though comes from a white paper that Ben Hindman and his team at UC Berkeley published back in 2010. Btw it is worth noting Hindman has moved on from Twitter to Mesosphere, a startup that is building and commercializing a data center OS with Mesos as the kernel. I will focus on distilling the main points of the white paper and then offering some thoughts on the value of the technology.
The Mesos Workflow
Le’s start by picking up where we left off in my previous post and…
View original post 1,564 more words
Infrastructure nerd learning to code and be a better human