spark技术分享1.0解释.pptx

Spark 技术分享;2;one stack to rule them all;BDAS(the Berkeley Data Analytics Stack);Spark Ecosystem;为什么比hadoop快;Performance;MapReduce Map;MapReduce Reduce;Shuffle;Basic Concepts;spark cluster overview;核心组件;Spark Runtime;调度（scheduler）;Spark的骨架—Akka;Spark的骨架—Akka;Spark的容错（节点级别）;Spark的容错（进程级别）;Spark Master HA;21;Resilient Distributed Dataset（RDD）弹性分布数据集;RDD的Partition;RDD Transfor Action;Lineage（血统）;Word Count Stage and RDD;Word Count Stage and RDD;Example-Word Count（Java version）;Example-Word Count（Scala version）;Code Size;31;Spark Streaming;Spark Streaming;DStream;VS Storm;Storm简介;Storm组件;Storm容错;39