Spark Core: sc.textFile vs sc.WholeTextFiles

「已注销」

于 2015-09-08 15:10:27 发布

阅读量3k

点赞数

分类专栏： spark

spark 专栏收录该内容

1 篇文章

订阅专栏

本文介绍了Spark中加载数据为RDD的两种不同方式：sc.textFile和sc.wholeTextFiles。前者将文件的每一行作为RDD的一个元素，而后者则将每个文件作为一个键值对，键为文件路径，值为文件的全部内容。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

While loading a RDD from source data, there are two choices which look similar.

scala&gt; val movies = sc.textFile("movies")

scala&gt; val movies = sc.wholeTextFiles("movies")

sc.textFile

SparkContext’s TextFile method, i.e., sc.textFile in Spark Shell, creates a RDD with each line as an element. If there are 10 files in movies folder, 10 partitions will be created. You can verify the number of partitions by:

scala&gt; movies.partitions.length

sc.wholeTextFiles

SparkContext’s whole text files method, i.e., sc.wholeTextFiles in Spark Shell, creates a PairRDD with the key being the file name with a path. It’s a full path like “hdfs://m1.zettabytes.com:9000/user/hduser/movies/movie1.txt”. The value is the whole content of file in String. Here the number of partitions will be 1 or more depending upon how many executor cores you have.