You're doing something wrong. I've used awk to run big data reformatting jobs in under an hour that took most of a day to run in Scala on an Apache Spark clust

You're doing something wrong. I've used awk to run big data reformatting jobs in under an hour that took most of a day to run in Scala on an Apache Spark cluster. In the vast majority of cases today, if speed is your problem, then you are the problem - especially since most problems fit into RAM these days, even w/o exotic stuff like RAMcloud...