The Big Data pipeline compound pattern generally comprises multiple stages whose objectives are to divide complex processing operations into down into modular steps for easier understanding and debugging and to be amenable to future data processing requirements.