A tool to monitor and manage sets of large high performance storage servers


June 2015-July 2017


Me, and about 10 other engineers at Oracle




Turns out it is useful to manage expensive servers in a unified tool


As documented in my more sentimental post on this, my choice to go work at Oracle for a while was a bit odd but ultimately rewarding. Working on a prototype and eventually an entire new product from scratch was a rare opportunity and a rewarding project. We prototyped a microservices driven concept on the side for about nine months with fancy new tools such as Docker and the various popular messaging queues and NoSQL DBs, before being given the green light to go ahead and develop the monitoring and management tool in Java as a proper product. In just about a year, we had finished a polished mostly-bug-free complex product that did a lot of data crunching and visualization. And we did this as a small team of about a dozen engineers, so I got to have broad and significant influence over the development of the backend. Even if I eventually left to go back to research (and the project was ultimately scrapped, after beta release), I think fondly of how we started with zero code and built a really solid management tool in just a year.