User`s guide

6 Using Database Toolbox Functions

6-86

Define the Mapper and Reducer Functions

You can write your own mapper function to process large data sets in chunks. This

example uses the mapper function meanArrivalDelayMapper.m. This mapper

function reads in arrival delay data from the DatabaseDatastore object, calculates

the number of delays and the total arrival delay in the chunk, and stores both values in

KeyValueStore. Display the code for this function using the command type.

type meanArrivalDelayMapper.m

function meanArrivalDelayMapper (data, info, intermKVStore)

% Mapper function for the MeanMapReduceExample.

% Data is an n-by-1 table of the ArrDelay. Remove missing value first:

data(isnan(data.ArrDelay),:) = [];

% Record the partial counts and sums and the reducer will accumulate them.

partCountSum = [length(data.ArrDelay), sum(data.ArrDelay)];

add(intermKVStore, 'PartialCountSumDelay',partCountSum);

Add this code at the beginning of the mapper function if you are using the native ODBC

interface to connect to the database.

if iscell(data)

return

end

This code skips the final output of the read function. The final output is a cell array

containing the string 'No Data'.

You can write your own reducer function to process large data sets in chunks. This

example uses the reducer function meanArrivalDelayReducer.m. This reducer

function reads in intermediate values for the number of delays and the total arrival

delay. Then, this function calculates the overall mean arrival delay. mapreduce calls this

reducer function once since the mapper function only adds one key to KeyValueStore.

Display the code for this function by using the command type.

type meanArrivalDelayReducer.m

function meanArrivalDelayReducer(intermKey, intermValIter, outKVStore)

% Reducer function for the MeanMapReduceExample.