[SPARK-21935] Pyspark UDF causing ExecutorLostFailure

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Incomplete
• Affects Version/s: 2.1.0
• Component/s: PySpark
• Labels: bulk-closed pyspark udf
• Assignee: Unassigned
• Reporter: Nikolaos Tsipas
• Created: 06/Sep/17 10:45
• Updated: 21/May/19 04:16
• Resolved: 21/May/19 04:16

Description

Hi,

I'm using spark 2.1.0 on AWS EMR (Yarn) and trying to use a UDF in python as follows:

from pyspark.sql.functions import col, udf
from pyspark.sql.types import StringType

path = 's3://some/parquet/dir/myfile.parquet'
df = spark.read.load(path)
def _test_udf(useragent):
    return useragent.upper()

test_udf = udf(_test_udf, StringType())
df = df.withColumn('test_field', test_udf(col('std_useragent')))
df.write.parquet('/output.parquet')

The following config is used in spark-defaults.conf (using maximizeResourceAllocation in EMR)

...
spark.executor.instances         4
spark.executor.cores             8
spark.driver.memory              8G
spark.executor.memory            9658M
spark.default.parallelism        64
spark.driver.maxResultSize       3G
...

The cluster has 4 worker nodes (+1 master) with the following specs: 8 vCPU, 15 GiB memory, 160 SSD GB storage

The above example fails every single time with errors like the following:

17/09/06 09:58:08 WARN TaskSetManager: Lost task 26.1 in stage 1.0 (TID 50, ip-172-31-7-125.eu-west-1.compute.internal, executor 10): ExecutorLostFailure (executor 10 exited caused by one of the running tasks) Reason: Container killed by YARN for exceeding memory limits. 10.4 GB of 10.4 GB physical memory used. Consider boosting spark.yarn.executor.memoryOverhead.

I tried to increase the spark.yarn.executor.memoryOverhead to 3000 which delays the errors but eventually I get them before the end of the job. The job eventually fails.

If I run the above job in scala everything works as expected (without having to adjust the memoryOverhead)

import org.apache.spark.sql.functions.udf

val upper: String => String = _.toUpperCase
val df = spark.read.load("s3://some/parquet/dir/myfile.parquet")
val upperUDF = udf(upper)
val newdf = df.withColumn("test_field", upperUDF(col("std_useragent")))
newdf.write.parquet("/output.parquet")

Cpu utilisation is very bad with pyspark

Is this a known bug with pyspark and udfs or is it a matter of bad configuration?
Looking forward to suggestions. Thanks!