Hi everyone, change data capture (CDC) database replication is so often misunderstood, with teams taking the choice between CDC and batch too lightly when it comes to analytics applications. I had a blast collaborating on this article with John Kutay of Striim about the true downstream positive impact of CDC as well as the technical implications to keep in mind. Take a read! https://towardsdatascience.com/when-change-data-capture-wins-271875e3df1a

In the spirit of Platform Engineering, I've also championed CDC as a way to support NoSQL-based solutions. I'm a big fan of cosmosDB for operational loads and often get resistance when teams need to make reporting-type queries. By building an event-based CDC into a data warehouse right away I've helped teams get the 'best of both worlds' Thanks for the article and I appreciate that graphic that shows OLTP vs OLAP