Geek Answers Handbook

Is MapReduce just a generalization of another programming principle?

I get parallel programming and I study mapreduce and other distributed algorithms. Is it best to study mapreduce or is there a more general algorithm that will serve me better?

+3
source share
4 answers

It depends on what you intend to use the algorithm for.

MapReduce is a generalized and very useful programming model. (Google bases many internal indexing processes on it). Studying this, of course, will not do you any harm.

MapReduce - . / / . , .

: - , , .

:

  • ,
+4

- , , .

Mapreduce - map a reduce. , .

, Scheme Clojure. " ", , .

+2

"" , MapReduce. , MapReduce, , , , , , MapReduce ( ).

+2

, , . (, pthreads), (, MPI MapReduce), .

MPI - . MPI, , MPI- , , "" , "" , . MapReduce - , MPI. , , , , . Hadoop MapReduce.

I think you better understand what MapReduce does and how you can implement this by writing some of your own MPI programs. You can't be sick to learn Hadoop, but when it comes to general knowledge of concurrent programming, it's good to be familiar with the basics such as pthreads, OpenMP, and MPI.

+1
source

More articles:


All Articles