Question

Given a JavaRDD<Double> myRDD, which holds a huge list of numbers. Our goal is to calculate the average of all these numbers. So the following code is being considered:

myRDD.reduce(New Functions2<Double, Double, Double>() {

public Double call(Double x, Double y) {

return (x+y)/2.0;

}

} );

2.1 (3 points) Select the ONLY ONE correct answer from the list given below:

1. this will work perfectly fine
2. this will not work as expected
3. this will work but can also result in overflow and loss of precision
4. this will work but will be slow

2.2 (3 points) Spark in fact provides an action call to implement the calculation of average value, and we have used it in one of our implementations. In which algorithm did we use this, and what is the name of the action call?

Answer 1

2.1)

The correct answer is:

b. this will not work as expected

2.2)

NOTE: The information provided is not sufficient to answer this part, hence skipping it.