I was trying to find that article someone wrote about how everyone doing microservices should have about 500 databases now. (actually I can't remember fully wha

I was trying to find that article someone wrote about how everyone doing microservices should have about 500 databases now. (actually I can't remember fully what that article was) But I was just hoping that whoever wrote it could provide details as it's clear the pendulum is definitely swinging back to monoliths. Also I feel like all tech goes this way. Years ago to do "big data" you had drill, kafka, HDFS, pick your cloudera or hortonworks, roll up your HBase, your storm, spark - hire a team to install it. It seems like now all we do is purchase Elastic cloud, and write a one-off spark script or pandas job and call it a f*cking night. I guess things go in circles.