There is a widening gap between the sophistication of manufacturing data models and the reality of the production line.
When Zaharia started work on Spark around 2010, analyzing "big data" generally meant using MapReduce, the Java-based ...