在我的HDFS中,我收集了大約350個csv文件。每個文件的大小從幾個KB到250Mb不等。我需要將這些csv文件的值插入到名爲RECORD的表中。插入時,我也需要引用一些其他表(PARAMETER和FRAME_RATE)。我有以下查詢來完成此任務。將記錄插入大量csv文件中的表格
-- create external table for the csv files in hdfs
CREATE EXTERNAL TABLE TEMP_CSV(
FRAME_RANK BIGINT,
FRATE BIGINT,
SOURCE STRING,
PARAM STRING,
RECORDEDVALUE STRING
)
ROW FORMAT DELIMITED
FIELDS TERMINATED BY ';'
location '/user/bala/output'
TBLPROPERTIES ("skip.header.line.count"="2");
-- Now insert fresh values into T_RECORD
INSERT OVERWRITE TABLE RECORD
PARTITION(SESSION)
SELECT DISTINCT
TEMP_CSV.F_FRAME_RANK,
PARAMETER.K_ID,
FRAME_RATE.K_ID,
CAST(TEMP_CSV.RECORDEDVALUE as FLOAT),
split(reverse(split(reverse(TEMP_CSV.INPUT__FILE__NAME),"/")[0]), "[.]")[0] AS SESSION
FROM TEMP_CSV , PARAMETER, FRAME_RATE
WHERE PARAMETER.NAME = TEMP_CSV.PARAM AND FRAME_RATE.FRATE = TEMP_CSV.FRATE;
在我小的PoC的研究,我有大約50 CSV文件,該查詢成功地填充記錄到記錄表中關於500秒與下面的配置
Hive-on-spark
spark standalon
6 nodes in the cluster
4 cores per node/16gb RAM
spark.executor.memory 2g
然而,當我處理350個文件,查詢失敗,執行程序中出現java堆空間錯誤。所以,我增加了executor.memory到4g。失敗。我增加了executor.memory到6g。失敗。最後,我增加了spark.executor.memory到12g。成功。但花了大約2小時30分鐘。將spark.executor.memory增加到12g導致每個節點只有一個執行器,因此只有6個執行器。
當我的executor.memory是6G,這是日誌在發生故障時,
******
******
2017-06-12 11:59:09,988 Stage-1_0: 101/101 Finished Stage-2_0: 12/12 Fini shed Stage-3_0: 0(+12,-2)/12
2017-06-12 11:59:12,997 Stage-1_0: 101/101 Finished Stage-2_0: 12/12 Finished Stage-3_0: 0(+12,-2)/12
2017-06-12 11:59:16,004 Stage-1_0: 101/101 Finished Stage-2_0: 12/12 Finished Stage-3_0: 0(+12,-2)/12
2017-06-12 11:59:19,012 Stage-1_0: 101/101 Finished Stage-2_0: 12/12 Finished Stage-3_0: 0(+12,-2)/12
*****
*****
在執行,這是錯誤日誌
17/06/12 11:58:36 WARN NettyRpcEndpointRef: Error sending message [message = Heartbeat(5,[Lscala.Tuple2;@e65f7b8,BlockManagerId(5, bndligpu04, 54618))] in 1 attempts
org.apache.spark.rpc.RpcTimeoutException: Futures timed out after [50 seconds]. This timeout is controlled by spark.executor.heartbeatInterval
at org.apache.spark.rpc.RpcTimeout.org$apache$spark$rpc$RpcTimeout$$createRpcTimeoutException(RpcTimeout.scala:48)
at org.apache.spark.rpc.RpcTimeout$$anonfun$addMessageIfTimeout$1.applyOrElse(RpcTimeout.scala:63)
at org.apache.spark.rpc.RpcTimeout$$anonfun$addMessageIfTimeout$1.applyOrElse(RpcTimeout.scala:59)
at scala.runtime.AbstractPartialFunction.apply(AbstractPartialFunction.scala:33)
at org.apache.spark.rpc.RpcTimeout.awaitResult(RpcTimeout.scala:76)
at org.apache.spark.rpc.RpcEndpointRef.askWithRetry(RpcEndpointRef.scala:101)
at org.apache.spark.executor.Executor.org$apache$spark$executor$Executor$$reportHeartBeat(Executor.scala:476)
at org.apache.spark.executor.Executor$$anon$1$$anonfun$run$1.apply$mcV$sp(Executor.scala:505)
at org.apache.spark.executor.Executor$$anon$1$$anonfun$run$1.apply(Executor.scala:505)
at org.apache.spark.executor.Executor$$anon$1$$anonfun$run$1.apply(Executor.scala:505)
at org.apache.spark.util.Utils$.logUncaughtExceptions(Utils.scala:1801)
at org.apache.spark.executor.Executor$$anon$1.run(Executor.scala:505)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:308)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:180)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:294)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
Caused by: java.util.concurrent.TimeoutException: Futures timed out after [50 seconds]
at scala.concurrent.impl.Promise$DefaultPromise.ready(Promise.scala:219)
at scala.concurrent.impl.Promise$DefaultPromise.result(Promise.scala:223)
at scala.concurrent.Await$$anonfun$result$1.apply(package.scala:107)
at scala.concurrent.BlockContext$DefaultBlockContext$.blockOn(BlockContext.scala:53)
at scala.concurrent.Await$.result(package.scala:107)
at org.apache.spark.rpc.RpcTimeout.awaitResult(RpcTimeout.scala:75)
... 14 more
17/06/12 11:58:36 ERROR Executor: Exception in task 0.0 in stage 3.0 (TID 115)
java.lang.OutOfMemoryError: Java heap space
at java.nio.HeapByteBuffer.<init>(HeapByteBuffer.java:57)
at java.nio.ByteBuffer.allocate(ByteBuffer.java:335)
at org.apache.orc.impl.OutStream.getNewInputBuffer(OutStream.java:109)
at org.apache.orc.impl.OutStream.write(OutStream.java:130)
at org.apache.orc.impl.RunLengthIntegerWriterV2.writeDeltaValues(RunLengthIntegerWriterV2.java:238)
at org.apache.orc.impl.RunLengthIntegerWriterV2.writeValues(RunLengthIntegerWriterV2.java:186)
at org.apache.orc.impl.RunLengthIntegerWriterV2.write(RunLengthIntegerWriterV2.java:772)
at org.apache.orc.impl.WriterImpl$IntegerTreeWriter.writeBatch(WriterImpl.java:1039)
at org.apache.orc.impl.WriterImpl$StructTreeWriter.writeRootBatch(WriterImpl.java:1977)
at org.apache.orc.impl.WriterImpl.addRowBatch(WriterImpl.java:2759)
at org.apache.hadoop.hive.ql.io.orc.WriterImpl.flushInternalBatch(WriterImpl.java:277)
at org.apache.hadoop.hive.ql.io.orc.WriterImpl.addRow(WriterImpl.java:296)
at org.apache.hadoop.hive.ql.io.orc.OrcOutputFormat$OrcRecordWriter.write(OrcOutputFormat.java:103)
at org.apache.hadoop.hive.ql.exec.FileSinkOperator.process(FileSinkOperator.java:743)
at org.apache.hadoop.hive.ql.exec.Operator.forward(Operator.java:837)
at org.apache.hadoop.hive.ql.exec.SelectOperator.process(SelectOperator.java:97)
at org.apache.hadoop.hive.ql.exec.spark.SparkReduceRecordHandler.processKeyValues(SparkReduceRecordHandler.java:309)
at org.apache.hadoop.hive.ql.exec.spark.SparkReduceRecordHandler.processRow(SparkReduceRecordHandler.java:267)
at org.apache.hadoop.hive.ql.exec.spark.HiveReduceFunctionResultList.processNextRecord(HiveReduceFunctionResultList.java:49)
at org.apache.hadoop.hive.ql.exec.spark.HiveReduceFunctionResultList.processNextRecord(HiveReduceFunctionResultList.java:28)
at org.apache.hadoop.hive.ql.exec.spark.HiveBaseFunctionResultList$ResultIterator.hasNext(HiveBaseFunctionResultList.java:95)
at scala.collection.convert.Wrappers$JIteratorWrapper.hasNext(Wrappers.scala:41)
at scala.collection.Iterator$class.foreach(Iterator.scala:727)
at scala.collection.AbstractIterator.foreach(Iterator.scala:1157)
at org.apache.spark.rdd.AsyncRDDActions$$anonfun$foreachAsync$1$$anonfun$apply$15.apply(AsyncRDDActions.scala:120)
at org.apache.spark.rdd.AsyncRDDActions$$anonfun$foreachAsync$1$$anonfun$apply$15.apply(AsyncRDDActions.scala:120)
at org.apache.spark.SparkContext$$anonfun$37.apply(SparkContext.scala:1992)
at org.apache.spark.SparkContext$$anonfun$37.apply(SparkContext.scala:1992)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:66)
at org.apache.spark.scheduler.Task.run(Task.scala:89)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:227)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
17/06/12 11:58:36 ERROR SparkUncaughtExceptionHandler: Uncaught exception in thread Thread[Executor task launch worker-1,5,main]
java.lang.OutOfMemoryError: Java heap space
at java.nio.HeapByteBuffer.<init>(HeapByteBuffer.java:57)
at java.nio.ByteBuffer.allocate(ByteBuffer.java:335)
at org.apache.orc.impl.OutStream.getNewInputBuffer(OutStream.java:109)
at org.apache.orc.impl.OutStream.write(OutStream.java:130)
at org.apache.orc.impl.RunLengthIntegerWriterV2.writeDeltaValues(RunLengthIntegerWriterV2.java:238)
我的問題是: -
- 是否有一個範圍來優化查詢?
- 任何其他可以應對這一挑戰的火花/蜂巢配置?
- 有沒有辦法告訴Hive處理50個文件?
任何幫助/信息來解決這個問題將是有益的。還有一個信息,'SELECT'聲明有效,我可以在我的色調瀏覽器中看到結果。當我嘗試插入由'SELECT'收集的信息時,查詢中斷。
感謝您的建議,但我不知道如何--executor-cores = 2將是正確的修復。因爲在一個核心中只有一個進程在6gb可用的情況下運行,它會失敗。再運行一個進程與6gb共享應該再次失敗。對?我可以嘗試一下。 – Bala