FelixYBW opened a new issue, #6527:
URL: https://github.com/apache/incubator-gluten/issues/6527
### Backend
VL (Velox)
### Bug description
the query plan:
```
Execute InsertIntoHadoopFsRelationCommand (21)
+- FakeRowAdaptor (20)
+- AdaptiveSparkPlan (19)
+- == Current Plan ==
ObjectHashAggregate (13)
+- ShuffleQueryStage (12)
+- ColumnarExchange (11)
+- ^ ProjectExecTransformer (9)
+- ^ InputIteratorTransformer (8)
+- ^ InputAdapter (7)
+- ^ RowToVeloxColumnar (6)
+- ^ ObjectHashAggregate (5)
+- ^ VeloxColumnarToRowExec (4)
+- ^ ProjectExecTransformer (2)
+- ^ Scan parquet
default.api_experiment_request_joins_v1 (1)
```
looks like there is an external sort in operator ObjectHashAggregate.
without PR6480, it reports OOM error:
```
org.apache.gluten.exception.GlutenException: java.lang.RuntimeException:
Exception: VeloxRuntimeError
Error Source: RUNTIME
Error Code: INVALID_STATE
Reason: Operator::getOutput failed for [operator: TableScan, plan node ID:
0]: Error during calling Java code from native code:
org.apache.gluten.memory.memtarget.ThrowOnOomMemoryTarget$OutOfMemoryException:
Not enough spark off-heap execution memory. Acquired: 8.0 MiB, granted: 0.0 B.
Try tweaking config option spark.memory.offHeap.size to get larger space to run
this application.
Current config settings:
spark.gluten.memory.offHeap.size.in.bytes=40.0 GiB
spark.gluten.memory.task.offHeap.size.in.bytes=10.0 GiB
spark.gluten.memory.conservative.task.offHeap.size.in.bytes=5.0 GiB
Memory consumer stats:
Task.1744:
Current used bytes: 10.0 GiB, peak bytes: N/A
+-
org.apache.spark.util.collection.unsafe.sort.UnsafeExternalSorter@265f7d0c:
Current used bytes: 9.8 GiB, peak bytes: N/A
+- Gluten.Tree.4:
Current used bytes: 192.0 MiB, peak bytes: 260.0 MiB
| \- root.4:
Current used bytes: 192.0 MiB, peak bytes: 260.0 MiB
| +- WholeStageIterator.4:
Current used bytes: 184.0 MiB, peak bytes: 192.0 MiB
| | \- single:
Current used bytes: 184.0 MiB, peak bytes: 192.0 MiB
| | +- task.Gluten_Stage_0_TID_1744:
Current used bytes: 184.0 MiB, peak bytes: 192.0 MiB
| | | +- node.0:
Current used bytes: 184.0 MiB, peak bytes: 192.0 MiB
| | | | +- op.0.0.0.TableScan:
Current used bytes: 184.0 MiB, peak bytes: 184.2 MiB
| | | | \- op.0.0.0.TableScan.test-hive:
Current used bytes: 0.0 B, peak bytes: 0.0 B
| | | \- node.1:
Current used bytes: 0.0 B, peak bytes: 0.0 B
| | | \- op.1.0.0.FilterProject:
Current used bytes: 0.0 B, peak bytes: 0.0 B
| | \- WholeStageIterator_default_leaf:
Current used bytes: 0.0 B, peak bytes: 0.0 B
| +- ColumnarToRow.4:
Current used bytes: 8.0 MiB, peak bytes: 8.0 MiB
| | \- single:
Current used bytes: 1024.0 KiB, peak bytes: 1024.0 KiB
| | \- ColumnarToRow_default_leaf:
Current used bytes: 512.0 KiB, peak bytes: 512.0 KiB
| +- OverAcquire.DummyTarget.8:
Current used bytes: 0.0 B, peak bytes: 57.6 MiB
| \- OverAcquire.DummyTarget.9:
Current used bytes: 0.0 B, peak bytes: 2.4 MiB
\-
org.apache.spark.util.collection.unsafe.sort.UnsafeExternalSorter@281f1779:
Current used bytes: 0.0 B, peak bytes: N/A
```
with PR6480, the error becomes:
```
4/07/20 04:21:35 ERROR [Thread-18] listener.ManagedReservationListener:
Error reserving memory from target
java.lang.IllegalStateException: Number of records written exceeded
numRecordsToWrite = 31956470
at
org.apache.spark.util.collection.unsafe.sort.UnsafeSorterSpillWriter.write(UnsafeSorterSpillWriter.java:118)
at
org.apache.spark.util.collection.unsafe.sort.UnsafeExternalSorter.spillIterator(UnsafeExternalSorter.java:623)
at
org.apache.spark.util.collection.unsafe.sort.UnsafeExternalSorter.forceSpill(UnsafeExternalSorter.java:234)
at
org.apache.spark.memory.TaskMemoryManager.acquireExecutionMemory(TaskMemoryManager.java:176)
at
org.apache.spark.memory.TaskMemoryManager.acquireExecutionMemory(TaskMemoryManager.java:125)
at
org.apache.spark.memory.MemoryConsumer.acquireMemory(MemoryConsumer.java:127)
at
org.apache.gluten.memory.memtarget.spark.TreeMemoryConsumer.borrow(TreeMemoryConsumer.java:65)
at
org.apache.gluten.memory.memtarget.TreeMemoryTargets$Node.borrow0(TreeMemoryTargets.java:126)
at
org.apache.gluten.memory.memtarget.TreeMemoryTargets$Node.borrow(TreeMemoryTargets.java:118)
at
org.apache.gluten.memory.memtarget.TreeMemoryTargets$Node.borrow0(TreeMemoryTargets.java:126)
at
org.apache.gluten.memory.memtarget.TreeMemoryTargets$Node.borrow(TreeMemoryTargets.java:118)
at
org.apache.gluten.memory.memtarget.OverAcquire.borrow(OverAcquire.java:62)
at
org.apache.gluten.memory.memtarget.ThrowOnOomMemoryTarget.borrow(ThrowOnOomMemoryTarget.java:35)
at
org.apache.gluten.memory.listener.ManagedReservationListener.reserve(ManagedReservationListener.java:43)
```
Looks PR6480 also triggered the sort in the external sort, but it has error.
@jinchengchenghh @zhztheplayer
### Spark version
Spark-3.2.x
### Spark configurations
_No response_
### System information
Velox System Info v0.0.2
Commit: 34dbec25d204fcb302893429350d37081feb5edf
CMake Version: 3.29.4
System: Linux-5.4.0-1063-aws
Arch: x86_64
CPU Name: Model name: Intel(R) Xeon(R) Platinum 8488C
C++ Compiler: /usr/bin/c++
C++ Compiler Version: 9.4.0
C Compiler: /usr/bin/cc
C Compiler Version: 9.4.0
CMake Prefix Path: /usr/local;/usr;/;/usr;/usr/local;/usr/X11R6;/usr/pkg;/opt
### Relevant logs
_No response_
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
