在spark2.0中，我有两个数据框架，我需要首先连接它们并执行reduceByKey来聚合数据。我总是在执行器中使用OOM。提前谢谢。
数据d1 (1G, 5亿行，缓存，由colid2分区)
id1 id2
1 1
1 3
1 4
2 0
2 7
...
d2 (160G, 200万行，缓存，按colid2分区，value col包含5000个浮点数列表)
id2 value
0 [0.1, 0.2, 0.0001, ...]
1 [0.001, 0.7, 0.0002, ...]
...
现在我需要连接两个表得到d3，我使用spark.sql
select d1.id1, d2.value
FROM d1 JOIN d2 ON d1.id2 = d2.id2
然后在d3上执行reduceByKey，并汇总表d1
中每个id1的值d4 = d3.rdd.reduceByKey(lambda x, y: numpy.add(x, y))
.mapValues(lambda x: (x / numpy.linalg.norm(x, 1)).toList)
.toDF()
我估计d4的大小为340G。现在我在r3.8xlarge机器上运行作业
mem: 244G
cpu: 64
Disk: 640G
我玩了一些配置，但我总是在执行器中得到OOM。问题是
是否可以在当前类型的机器上运行此作业?或者我应该用更大的机器(多大?)但我记得我看到过一些文章/博客说用相对较小的机器进行tb级处理。
我应该做什么样的改进?例如spark配置，代码优化?
是否有可能估计每个执行器所需的内存量?
我尝试了一些Spark配置
config1:
--verbose
--conf spark.sql.shuffle.partitions=200
--conf spark.dynamicAllocation.enabled=false
--conf spark.driver.maxResultSize=24G
--conf spark.shuffle.blockTransferService=nio
--conf spark.serializer=org.apache.spark.serializer.KryoSerializer
--conf spark.kryoserializer.buffer.max=2000M
--conf spark.rpc.message.maxSize=800
--conf "spark.executor.extraJavaOptions=-verbose:gc - XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:MetaspaceSize=100M"
--num-executors 4
--executor-memory 48G
--executor-cores 15
--driver-memory 24G
--driver-cores 3
config2:
--verbose
--conf spark.sql.shuffle.partitions=10000
--conf spark.dynamicAllocation.enabled=false
--conf spark.driver.maxResultSize=24G
--conf spark.shuffle.blockTransferService=nio
--conf spark.serializer=org.apache.spark.serializer.KryoSerializer
--conf spark.kryoserializer.buffer.max=2000M
--conf spark.rpc.message.maxSize=800
--conf "spark.executor.extraJavaOptions=-verbose:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:MetaspaceSize=100M"
--num-executors 4
--executor-memory 48G
--executor-cores 15
--driver-memory 24G
--driver-cores 3
--verbose
--conf spark.sql.shuffle.partitions=10000
--conf spark.dynamicAllocation.enabled=true
--conf spark.driver.maxResultSize=6G
--conf spark.shuffle.blockTransferService=nio
--conf spark.serializer=org.apache.spark.serializer.KryoSerializer
--conf spark.kryoserializer.buffer.max=2000M
--conf spark.rpc.message.maxSize=800
--conf "spark.executor.extraJavaOptions=-verbose:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:MetaspaceSize=100M"
--executor-memory 6G
--executor-cores 2
--driver-memory 6G
--driver-cores 3
--verbose
--conf spark.sql.shuffle.partitions=20000
--conf spark.dynamicAllocation.enabled=false
--conf spark.driver.maxResultSize=6G
--conf spark.shuffle.blockTransferService=nio
--conf spark.serializer=org.apache.spark.serializer.KryoSerializer
--conf spark.kryoserializer.buffer.max=2000M
--conf spark.rpc.message.maxSize=800
--conf "spark.executor.extraJavaOptions=-verbose:gc -XX:+PrintGCDetails -XX:+PrintGCTimeStamps -XX:MetaspaceSize=100M"
--num-executors 13
--executor-memory 15G
--executor-cores 5
--driver-memory 13G
--driver-cores 5
OOM Error1 from executor
ExecutorLostFailure (executor 14 exited caused by one of the running tasks) Reason: Container killed by YARN for exceeding memory limits. 9.1 GB of 9 GB physical memory used. Consider boosting spark.yarn.executor.memoryOverhead.
Heap
PSYoungGen total 1830400K, used 1401721K [0x0000000740000000, 0x00000007be900000, 0x00000007c0000000)
eden space 1588736K, 84% used [0x0000000740000000,0x0000000791e86980,0x00000007a0f80000)
from space 241664K, 24% used [0x00000007af600000,0x00000007b3057de8,0x00000007be200000)
to space 236032K, 0% used [0x00000007a0f80000,0x00000007a0f80000,0x00000007af600000)
ParOldGen total 4194304K, used 4075884K [0x0000000640000000, 0x0000000740000000, 0x0000000740000000)
object space 4194304K, 97% used [0x0000000640000000,0x0000000738c5b198,0x0000000740000000)
Metaspace used 59721K, capacity 60782K, committed 61056K, reserved 1101824K
class space used 7421K, capacity 7742K, committed 7808K, reserved 1048576K
OOM Error2 from executor
ExecutorLostFailure (executor 7 exited caused by one of the running tasks) Reason: Container marked as failed: container_1477662810360_0002_01_000008 on host: ip-172-18-9-130.ec2.internal. Exit status: 52. Diagnostics: Exception from container-launch.
Heap
PSYoungGen total 1968128K, used 1900544K [0x0000000740000000, 0x00000007c0000000, 0x00000007c0000000)
eden space 1900544K, 100% used [0x0000000740000000,0x00000007b4000000,0x00000007b4000000)
from space 67584K, 0% used [0x00000007b4000000,0x00000007b4000000,0x00000007b8200000)
to space 103936K, 0% used [0x00000007b9a80000,0x00000007b9a80000,0x00000007c0000000)
ParOldGen total 4194304K, used 4194183K [0x0000000640000000, 0x0000000740000000, 0x0000000740000000)
object space 4194304K, 99% used [0x0000000640000000,0x000000073ffe1f38,0x0000000740000000)
Metaspace used 59001K, capacity 59492K, committed 61056K, reserved 1101824K
class space used 7300K, capacity 7491K, committed 7808K, reserved 1048576K
容器出错
16/10/28 14:33:21 ERROR CoarseGrainedExecutorBackend: RECEIVED SIGNAL TERM
16/10/28 14:33:26 ERROR Utils: Uncaught exception in thread stdout writer for python
java.lang.OutOfMemoryError: Java heap space
at org.apache.spark.sql.catalyst.expressions.UnsafeRow.copy(UnsafeRow.java:504)
at org.apache.spark.sql.catalyst.expressions.GeneratedClass$GeneratedIterator.processNext(Unknown Source)
at org.apache.spark.sql.execution.BufferedRowIterator.hasNext(BufferedRowIterator.java:43)
at org.apache.spark.sql.execution.WholeStageCodegenExec$$anonfun$doExecute$3$$anon$2.hasNext(WholeStageCodegenExec.scala:386)
at scala.collection.Iterator$$anon$11.hasNext(Iterator.scala:408)
at org.apache.spark.api.python.SerDeUtil$AutoBatchedPickler.next(SerDeUtil.scala:120)
at org.apache.spark.api.python.SerDeUtil$AutoBatchedPickler.next(SerDeUtil.scala:112)
at scala.collection.Iterator$class.foreach(Iterator.scala:893)
at org.apache.spark.api.python.SerDeUtil$AutoBatchedPickler.foreach(SerDeUtil.scala:112)
at org.apache.spark.api.python.PythonRDD$.writeIteratorToStream(PythonRDD.scala:504)
at org.apache.spark.api.python.PythonRunner$WriterThread$$anonfun$run$3.apply(PythonRDD.scala:328)
at org.apache.spark.util.Utils$.logUncaughtExceptions(Utils.scala:1877)
at org.apache.spark.api.python.PythonRunner$WriterThread.run(PythonRDD.scala:269)
16/10/28 14:33:36 ERROR Utils: Uncaught exception in thread driver-heartbeater
16/10/28 14:33:26 ERROR Utils: Uncaught exception in thread stdout writer for python
java.lang.OutOfMemoryError: GC overhead limit exceeded
at java.lang.Double.valueOf(Double.java:519)
at org.apache.spark.sql.catalyst.expressions.UnsafeArrayData.get(UnsafeArrayData.java:138)
at org.apache.spark.sql.catalyst.util.ArrayData.foreach(ArrayData.scala:135)
at org.apache.spark.sql.execution.python.EvaluatePython$.toJava(EvaluatePython.scala:64)
at org.apache.spark.sql.execution.python.EvaluatePython$.toJava(EvaluatePython.scala:57)
at org.apache.spark.sql.Dataset$$anonfun$55.apply(Dataset.scala:2517)
at org.apache.spark.sql.Dataset$$anonfun$55.apply(Dataset.scala:2517)
at scala.collection.Iterator$$anon$11.next(Iterator.scala:409)
at org.apache.spark.api.python.SerDeUtil$AutoBatchedPickler.next(SerDeUtil.scala:121)
at org.apache.spark.api.python.SerDeUtil$AutoBatchedPickler.next(SerDeUtil.scala:112)
at scala.collection.Iterator$class.foreach(Iterator.scala:893)
at org.apache.spark.api.python.SerDeUtil$AutoBatchedPickler.foreach(SerDeUtil.scala:112)
at org.apache.spark.api.python.PythonRDD$.writeIteratorToStream(PythonRDD.scala:504)
at org.apache.spark.api.python.PythonRunner$WriterThread$$anonfun$run$3.apply(PythonRDD.scala:328)
at org.apache.spark.util.Utils$.logUncaughtExceptions(Utils.scala:1877)
at org.apache.spark.api.python.PythonRunner$WriterThread.run(PythonRDD.scala:269)
16/10/28 14:33:43 ERROR SparkUncaughtExceptionHandler: [Container in shutdown] Uncaught exception in thread Thread[stdout writer for python,5,main]
如果我按id2划分，看起来数据d1是非常倾斜的。因此，连接将导致OOM。如果d1像我之前认为的那样均匀分布，那么上面的配置应该是有效的。
<标题>更新2 h1> 贴出了我解决这个问题的方法，以防有人也遇到类似的问题。Attempt1
我的问题是，如果我划分d1除以id2，那么数据是相当偏斜的。因此，存在一些包含几乎所有id1的分区。因此，与d2的JOIN将导致OOM错误。为了缓解这样的问题，我首先从id2中识别一个子集s，如果按id2分区，它可能会导致这样的倾斜数据。然后我从d2创建d5，只包括s，从d2创建d6，不包括s。幸运的是，d5的尺寸不是太大。我可以广播连接d1和d5。然后连接d1和d6。然后，合并两个结果并执行reduceByKey操作。我马上就要解决这个问题了。我没有继续这样做，因为d1以后会变得更大。换句话说，这种方法对我来说是不可扩展的
Attempt2
幸运的是，在我的例子中，d2中的大多数值都很小。根据我的应用程序，我可以安全地删除小值并将向量转换为sparseVector，以显着减少d2的大小。在此之后，我将d1划分为id1，并广播连接d2(在删除小值之后)。当然，必须提高驱动程序内存以允许相对较大的广播变量。
你可以试试:把执行器的大小减小一点。你现在有:
--executor-memory 48G
--executor-cores 15
试一试:
--executor-memory 16G
--executor-cores 5
由于各种原因，较小的执行器大小似乎是最佳的。其中之一是java堆大小大于32G会导致对象引用从4个字节变为8个字节，并且所有内存需求都会激增。
编辑:问题实际上可能是d4分区太大(尽管其他建议仍然适用!)。您可以通过将d3重新划分为更大数量的分区(大致为d1 * 4)来解决这个问题，或者将其传递给reduceByKey的numPartitions可选参数。这两个选项都将触发洗牌，但这比崩溃要好。
我也遇到了同样的问题，但是我找了很多答案都不能解决我的问题。最后，我一步一步地调试我的代码。我发现由于每个分区的数据大小不平衡而导致的问题。点击df_rdd.repartition(nums)