You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@gluten.apache.org by "zhztheplayer (via GitHub)" <gi...@apache.org> on 2025/01/10 08:28:08 UTC
[I] [VL] Use of columnar table cache causes error when data emitted from the cached plan is in vanilla columnar format [incubator-gluten]
zhztheplayer opened a new issue, #8497:
URL: https://github.com/apache/incubator-gluten/issues/8497
### Backend
VL (Velox)
### Bug description
The [current code](https://github.com/apache/incubator-gluten/blob/6d71d27884dd6bda38848cb012a63fb37ee2d298/backends-velox/src/main/scala/org/apache/spark/sql/execution/ColumnarCachedBatchSerializer.scala#L103-L117) for determining whether to use columnar-based cache doesn't cover all possible cases. For example, the case when a vanilla Spark columnar scan fell back by unsupported data type.
If the case is met, the following error will be thrown:
```
org.apache.spark.SparkException: Job aborted due to stage failure: Task 0 in stage 10.0 failed 1 times, most recent failure: Lost task 0.0 in stage 10.0 (TID 10) (55f8ad28c275 executor driver): java.lang.IllegalStateException: Heavy batch should consist of arrow vectors
at org.apache.gluten.columnarbatch.ColumnarBatches.identifyBatchType(ColumnarBatches.java:81)
at org.apache.gluten.columnarbatch.ColumnarBatches.isLightBatch(ColumnarBatches.java:99)
at org.apache.gluten.columnarbatch.ColumnarBatches.ensureOffloaded(ColumnarBatches.java:134)
at org.apache.gluten.columnarbatch.VeloxColumnarBatches.ensureVeloxBatch(VeloxColumnarBatches.java:104)
at org.apache.spark.sql.execution.ColumnarCachedBatchSerializer.$anonfun$convertColumnarBatchToCachedBatch$2(ColumnarCachedBatchSerializer.scala:177)
at scala.collection.Iterator$$anon$9.next(Iterator.scala:577)
at org.apache.spark.sql.execution.ColumnarCachedBatchSerializer$$anon$1.next(ColumnarCachedBatchSerializer.scala:183)
at org.apache.spark.sql.execution.ColumnarCachedBatchSerializer$$anon$1.next(ColumnarCachedBatchSerializer.scala:179)
at scala.collection.Iterator$$anon$9.next(Iterator.scala:577)
at org.apache.spark.storage.memory.MemoryStore.putIterator(MemoryStore.scala:224)
at org.apache.spark.storage.memory.MemoryStore.putIteratorAsValues(MemoryStore.scala:302)
at org.apache.spark.storage.BlockManager.$anonfun$doPutIterator$1(BlockManager.scala:1531)
at org.apache.spark.storage.BlockManager.org$apache$spark$storage$BlockManager$$doPut(BlockManager.scala:1458)
at org.apache.spark.storage.BlockManager.doPutIterator(BlockManager.scala:1522)
at org.apache.spark.storage.BlockManager.getOrElseUpdate(BlockManager.scala:1349)
at org.apache.spark.rdd.RDD.getOrCompute(RDD.scala:378)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:329)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:367)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:331)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:367)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:331)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:367)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:331)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:367)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:331)
at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:52)
at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:367)
at org.apache.spark.rdd.RDD.iterator(RDD.scala:331)
at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:92)
at org.apache.spark.TaskContext.runTaskWithListeners(TaskContext.scala:161)
at org.apache.spark.scheduler.Task.run(Task.scala:139)
at org.apache.spark.executor.Executor$TaskRunner.$anonfun$run$3(Executor.scala:554)
at org.apache.spark.util.Utils$.tryWithSafeFinally(Utils.scala:1529)
at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:557)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:750)
```
### Spark version
None
### Spark configurations
_No response_
### System information
_No response_
### 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: commits-unsubscribe@gluten.apache.org.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@gluten.apache.org
For additional commands, e-mail: commits-help@gluten.apache.org
Re: [I] [VL] Use of columnar table cache causes error when data emitted from the cached plan is in vanilla columnar format [gluten]
Posted by "zhztheplayer (via GitHub)" <gi...@apache.org>.
zhztheplayer closed issue #8497: [VL] Use of columnar table cache causes error when data emitted from the cached plan is in vanilla columnar format
URL: https://github.com/apache/gluten/issues/8497
--
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: commits-unsubscribe@gluten.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@gluten.apache.org
For additional commands, e-mail: commits-help@gluten.apache.org
Re: [I] [VL] Use of columnar table cache causes error when data emitted from the cached plan is in vanilla columnar format [incubator-gluten]
Posted by "zhztheplayer (via GitHub)" <gi...@apache.org>.
zhztheplayer commented on issue #8497:
URL: https://github.com/apache/incubator-gluten/issues/8497#issuecomment-2586107568
Plan to fix the issue:
1. Add test case https://github.com/apache/incubator-gluten/pull/8498
2. Add a query planer hook similar to AdaptiveContext.isAdaptiveContext, to determine whether the Gluten columnar rule is called for columnar table cache's inner query plan's optimization
3. Unifi the hook API for columnar table cache and AQE
4. Fix test cases
--
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: commits-unsubscribe@gluten.apache.org
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@gluten.apache.org
For additional commands, e-mail: commits-help@gluten.apache.org