iceberg snapshots affect storage footprint (not performance)

it is easy to understand why most folks initially imagine that iceberg’s ability to maintain a long history of snapshots will cause performance problems, but that is not the case — the real gotcha is that keeping many versions can quickly consume 2-10+ times the amount of data lake storage space

becoming a data engineer (yet another top 10 list)

after a recent class i was asked what skills someone needs to become a data engineer – there are plenty of these lists all over the internet, yet here i go assuming i know enough to jot down yet another; at least i put mine all in a single picture 😉

updated streaming supervision features scorecard (added flink)

added apache flink to the comparison grid of kafka streams, spark streaming, and storm focused on the features they offer the operations side of the devops formula — it measures up well

batch as a “special case” of flink streaming (yes, now we’re mv’ing streaming back to batch)

the third part of a loosely coupled trilogy on flink batch and streaming that take us full-circle with the collapse of the DataSet API into the DataStream API — i’m not sure Run-D.M.C. could make this less tricky

big data api’s look a lot alike (code comparison with flink, kafka, spark, trident and pig)

exploring the similarity of the APIs from flink, kafka streams, spark (RDDs & DFs), storm’s trident and yes, even good old pig by implementing the canonical word count solution with each framework

topology supervision features of streaming frameworks (or lack thereof)

a smackdown of sort pitting kafka streams, spark streaming, and storm against each other — not for the features they give developers, but for the features they offer the operations side of the devops formula