摘要:
入门:知道MapReduce大致流程,map, shuffle, reduce知道combiner, partition作用,设置compression搭建hadoop集群,master/slave 都运行那些服务HDFS,replica如何定位版本0.20.2->0.20.203->0.20.205, 0.21, 0.23, 1.0. 1新旧API不同进阶:.Hadoop 参数调优,cluster level: JVM, map/reduce slots, job level: reducer #,memory, use combiner? use compression?pig 阅读全文