关键词: Hadoop, MapReduce, Shuffle, 性能

Abstract: As the core component of Hadoop, MapReduce is a programming model, It determines the performance and efficiency of Hadoop in its treatment of large data. Putting forward this idea that combining a large amount of the temporary result data produced by many Map tasks in the same job of the Map node for a lot of time consuming by Reduce end pulling a large amount of result data from the Map end via the internet,replaceing the mechanism that the original MapReduce architecture combines the result data of a single Map task .Through the improved project, the amount of output result data decreased on the Map node ,so that the amount of data transmission of the entire cluster is decreased ,saving the time of Reduce end copying Map end output data ,so the execution time of the MapReuce job is reduced, which improves the execution performance of the MapReduce .

Key words: Hadoop, MapReduce, Shuffle, performance