Skip to content

README.md

Latest commit

 

History

History
25 lines (19 loc) · 1.6 KB

README.md

File metadata and controls

25 lines (19 loc) · 1.6 KB

Distributed implementation of MapReduce

A distributed implementation of the MapReduce programming model in GoLang using a word count application as an example. The term "Coordinator" has been used in the implementation instead of the paper's "Master".

Go setup

Follow the go installation guide and check if it's correctly installed using go version

src/main directory

It contains mrcoordinator.go (the main program that creates a coordinator), mrworker.go (the main program that creates a worker) & text files to be used for word count.

src/mr directory

It contains the distributed part of our implementation in the form of coordinator.go, worker.go & rpc.go

src/mrapps directory

wc.go is the word count application containing the Map() & Reduce() function calls

Running MapReduce

  • git clone https://github.com/ehtesham0337/MapReduce.git
  • Open 2 separate terminals
  • Navigate to the /src/main directory in both
  • Build the word count plugin using go build -buildmode=plugin ../mrapps/wc.go
  • Run the Coordinator using go run mrcoordinator.go pg*.txt
  • In the 2nd terminal, run the worker (you may run more than 1 worker in separate terminals) using go run mrworker.go wc.so
  • Files mr-{0-7}-{0-9} in src/main will have all Key/Value pairs with the word as key and its occurence as value
  • Files mr-out-{0-9} contain the output with a list of words followed by the total number of counts