减少partition时,用coalesce效率更高
测试
repartition,shuffle 2.8G, 耗时10min39sec
df.rdd.repartition(1).saveAsTextFile("/gx/gziptest", classOf[org.apache.hadoop.io.compress.GzipCodec])
joe: start time: Tue Jul 07 12:43:06 CST 2015
joe: end time: Tue Jul 07 12:53:45 CST 2015
coalesce,没有shuffle, 耗时6min22sec
df.rdd.coalesce(1).saveAsTextFile("/gx/gziptest", classOf[org.apache.hadoop.io.compress.GzipCodec])
joe: start time: Tue Jul 07 13:39:16 CST 2015
joe: end time: Tue Jul 07 13:45:38 CST 2015
说明
repartition(numPartitions)
Reshuffle the data in the RDD randomly to create either more or fewer partitions and balance it across them. This always shuffles all data over the network.
等于
coalesce(numPartitions, shuffle = true)
coalesce(numPartitions)
Decrease the number of partitions in the RDD to numPartitions. Useful for running operations more efficiently after filtering down a large dataset.
If you are decreasing the number of partitions in this RDD, consider using `coalesce`,
which can avoid performing a shuffle.
However, if you're doing a drastic coalesce, e.g. to numPartitions = 1,
this may result in your computation taking place on fewer nodes than
you like (e.g. one node in the case of numPartitions = 1). To avoid this,
you can pass shuffle = true. This will add a shuffle step, but means the
current upstream partitions will be executed in parallel (per whatever
the current partitioning is).