305Big Data Processing with MapReduce
tion of how the Hadoop project is organized, see section “Open-Source
Implementation: Hadoop.”
Disco
Disco is another open source implementation of the MapReduce program-
ming model, developed at Nokia Research Center as a lightweight frame-
work for rapid scripting of distributed data processing.
The Disco core is written in Erlang, a functional language that is designed
for building robust fault-tolerant distributed applications. MapReduce
programs are typically written in Python, though, which lowers the entry
barrier and makes it possible to write data-processing code in only a few
lines of code.
Unlike Hadoop, Disco is only a minimal MapReduce implementation, and
does not include a customized ...