Questions tagged [MapReduce] (17)

I want to write (more precisely already wrote, I want to publish) a post on the topic of MapReduce 2.0 / YARN. I wonder whether this topic will be Hebraist? To facilitate the answer to the question will lead the content of the post: Introduction (the actuality, the history of the development of Hadoop MapReduce 2.0) 1. Ha...

Hello! Is there an example implementation of MapReduce without Hadoop? If in the context of the problem, that(example): There are N database servers with the same table structure with different data set. I want to send a request to all N servers and get one answer. Thanks in advance!

The point is this: there are enough powerful hardware, 128gb, 2 x Xeon E5-2620, Ryde on 3 TB. All this is bare Linux and mongodb 2.4.10, the base is big, about 320GB, 100 million records. You are only recording and constantly run mapreduce processes. It is clear that the Mr processes are not yuzayut all 2 x 2 x 4 cores.. so...

Hello, I use mongodb + built-in implementation of mapreduce. The data is as follows: { obj: integer, likes: integer}, ... I need to get 2 documents for each unique type of obj with maximum number of likes. Let's have the following data:1=>124, 1=>20, 1=>70, 1=>150, 3 => 500, 3=>499, 3=>0 As a result, I...

Make course, want to investigate the acceleration of computing "wordcount" to dump Wikipedia(600MB). Before each new run often add a new node($HADOOP_HOME/conf/slaves) OS: ubuntu server 12.04.04, hadoop-1.2.1

Greetings to all! Gather information for the review of software libraries and frameworks that are frequently used to solve their customer or high-performance tasks. I would be very grateful to all who share information, problems to be solved and with the help of the software.

I have a set of elements that can have a parent of the same type. Elements have parents, form graphs (connected lists). My task is to for each item, with parent a vertex to be specified as the parent element, which is the peak of the whole chain. For example: from 1 -> 2 -> 3 -> 4 you need to: 1 -> 4, 2 -> 4,...

Help to understand with MapReduce, I started reading about it has added Hadoop, and others. In General, a lot of articles where it is told about map() and reduce () but no where does it say that is a pair key-value.How to choose them or to understand that this model does not fit here. And is it possible for non coherent dat...

Or like this image to it not using Hadoop?

Advise where to start to learn Hadoop & MapReduce? (So far I found only "Hadoop: The Definitive Guide, 3rd Edition")