I jumped into backend programming in 2005. Back then there was one big database server, and we had load balancing with IP tables… load balancing based on the last digit of the client IP… the face of my friend, Peter Molnar, was priceless, when he saw that. During the years I saw memcached and database replication being introduced to our systems. Even NoSQL started to appear… however I never felt like it was my thing. I liked good old SQL more. The trend was too fancy for me back then. Nowadays I work with AWS and DynamoDB, but MySQL is still around, even Amazon knows that.
Today I came across this awesome article: Cache is the New RAM by icon-twitter Carlos Bueno which is a transcript of the talk from Defrag 2014. In it he takes us on a tour about maturing of fast storage solutions for the web, with an interesting conclusion. To tie this back to where I started, I would like to quote one paragraph from the article:
To paraphrase Paul Graham’s unbearably smug comment about Lisp: once you add group by, filter & join, you can no longer claim to have invented a new query language, only a new dialect of SQL. With worse syntax and no optimizer.
I feel like skipping the NoSQL bandwagon wasn’t a mistake. I strongly recommend reading this great overview of the last 15 years and a little sneak peek into the future.