I’m referring to Google’s 2004 MapReduce paper. The functional style that MapReduce derives from had been used in parallel computing models, e.g. the scatter/c

I’m referring to Google’s 2004 MapReduce paper. The functional style that MapReduce derives from had been used in parallel computing models, e.g. the scatter/compute/gather model of MPI, and in turn this was adopted by Hadoop, CouchDB, MongoDB and others.