It was an epic failure because you need a team to support and guide Cassandra use properly but no one wanted to do the grunt work. The VP of infrastructure MM o

It was an epic failure because you need a team to support and guide Cassandra use properly but no one wanted to do the grunt work. The VP of infrastructure MM openly called it “toil vs talent”, meaning those that did the grunt work would be held in high esteem and get yearly bonuses, but the promotions would go to those with “talent”, ie creating new things. When people are openly and stupidly incentivized like this, expect those people to behave in a predictable way. People started building new services to get promotions instead of “toiling” at supporting their fellow engineers. It affected most of engineering but especially in teams like Cassandra, where you needed guidance and support to properly use it effectively, it was a disaster. There should have been open office hours to help people with questions and to ensure that teams were using it properly but there wasn’t. Instead people were left to do what they wanted with no structure or guidance and Cassandra was completely misused. Productions problems ensued, people left the team because they didn’t want to be oncall fixing fires all the time, and eventually it came to the point where they decided to stop supporting it altogether. It was a complete disaster caused by very poor engineering management. We all knew that Netflix and Facebook use it without issues, but because of stupid management, it failed at Uber.