Data Mining and Big Data by Ying Tan Hideyuki Takagi & Yuhui Shi

Author:Ying Tan, Hideyuki Takagi & Yuhui Shi
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

2.2 Execution Stage

From the PipePlan stage, as showed in Fig. 2, there are four MR jobs. MR #1, MR#2 and MR#3 are all from Algorithm 2. In execution stage, there are two techniques: MapReduce jobs and multipipe procedure. The details in MapReduce and MultiPipeMap are showed in Algorithm 2–4. And the most important challenge is how to aggregate the skyline points through the MapReduce framework. Since the pipe plan is built, the MapReduce procedure will be launched. Algorithm 2,3 describe the MapReduce procedure, which are referred as MR#1-3 in Fig. 2. In the MapReduce procedure, The Map procedure loads local data according to sort tree, and emit key-value pair, in which cuboid as key and LSP as value. The LSP is produced by The skylineAggregation procedure, whose details are explained below. The Reduce procedure combines the skyline points in the same cuboid, and outputs the final skyline points set.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.