摘要
MapReduce是由并行编程模型及相关支撑系统组成的数据处理框架,通过定义接口和运行时支持库,通过定义良好的接口和运行时支持库,能够自动并行执行大规模计算任务,通过隐藏底层实现细节,降低实现并行编程的难度,Hadoop是目前MapReduce框架最流行的开源实现。文章首先介绍了MapReduce并行编程模型及其hadoop的运行原理、运行机制,深入研究了MapReduce计算任务在Hadoop系统中的运行过程。
MapReduce is composed of parallel programming model and its support system data processing framework,through the definition of interface support library and runtime support library,through the definition of a good interface and operation,capable of automatic parallel execution of large-scale computing tasks,by hiding the underlying implementation details,reduce the difficulty of parallel programming,Hadoop is currently the most popular MapReduce framework open source implementation.Firstly,this paper introduces the MapReduce parallel programming model and the operation principle and operation mechanism of Hadoop,and deeply studies the operation process of MapReduce computing task in Hadoop system.
出处
《电子测试》
2017年第9期77-78,共2页
Electronic Test
