Minimal Map Reduce $(10)$ Suppose we wish to count the number of characters in text

files on a system of three nodes $($Alpha$,$ Beta, Gamma$),$ and we wish to use the Map Reduce paradigm. For

the sake of this example, we will assume the following distribution of the data across the nodes:

Node File Contents

Alpha abcdabcdabcd

Beta axyaxyaxy

Gamma adyadyady

$1.$ Write a pseudo code implementation of a Map and Reduce function to calculate the character count

$($see Dean et al$.$ for inspiration$).(2)$

$2.$ Provide the output of the map tasks and how the output is distributed over nodes $($e$.$g$.$ Node$1$: $($key$1,$

value$1),($key$2,$ value$2),$ Node$2$: $($key$3,$ value$3)...).(2)$

$3.$ Suppose you have two reducer processes $($say$,$ R$1$ and R$2)$ that will execute your custom reducer

code, which data will be processed at each reducer $($e$.$g$.$ R$1$: $($key$2,$ value$2),$ R$2$: $($key$3,$ value$3),($key$1,$

value$1)))?(2)$

$4.$ During this entire process, which data are written to disk? $(2)$

$5.$ You wish to reduce network traffic between the mappers and the reducers using an additional stage,

what is this stage called and how would it act on the data in this example? Demonstrate this by

modifying the key$-$value pairs as necessary. $(2)$