package shims
- Alphabetic
- Public
- All
Type Members
- case class GpuBatchScanExec(output: Seq[AttributeReference], scan: Scan, runtimeFilters: Seq[Expression] = Seq.empty) extends SparkPlan with DataSourceV2ScanExecBase with GpuBatchScanExecMetrics with Product with Serializable
- class GpuCustomShuffleReaderMeta extends SparkPlanMeta[AQEShuffleReadExec]
-
class
GpuDataSourceRDD extends DataSourceRDD
A replacement for DataSourceRDD that does NOT compute the bytes read input metric.
A replacement for DataSourceRDD that does NOT compute the bytes read input metric. DataSourceRDD assumes all reads occur on the task thread, and some GPU input sources use multithreaded readers that cannot generate proper metrics with DataSourceRDD.
- Note
It is the responsibility of users of this RDD to generate the bytes read input metric explicitly!
- trait GpuDeterministicFirstLastCollectShim extends Expression
- case class GpuHashPartitioning(expressions: Seq[Expression], numPartitions: Int) extends GpuHashPartitioningBase with Product with Serializable
-
case class
GpuRangePartitioning(gpuOrdering: Seq[SortOrder], numPartitions: Int) extends Expression with GpuExpression with ShimExpression with GpuPartitioning with Product with Serializable
A GPU accelerated
org.apache.spark.sql.catalyst.plans.physical.Partitioningthat partitions sortable records by range into roughly equal ranges.A GPU accelerated
org.apache.spark.sql.catalyst.plans.physical.Partitioningthat partitions sortable records by range into roughly equal ranges. The ranges are determined by sampling the content of the RDD passed in.- Note
The actual number of partitions created might not be the same as the
numPartitionsparameter, in the case where the number of sampled records is less than the value ofpartitions. The GpuRangePartitioner is where all of the processing actually happens.
- class GpuSpecifiedWindowFrameMeta extends GpuSpecifiedWindowFrameMetaBase
- class GpuWindowExpressionMeta extends GpuWindowExpressionMetaBase
- case class GpuWindowInPandasExec(windowExpression: Seq[Expression], gpuPartitionSpec: Seq[Expression], cpuOrderSpec: Seq[SortOrder], child: SparkPlan)(cpuPartitionSpec: Seq[Expression]) extends SparkPlan with GpuWindowInPandasExecBase with Product with Serializable
-
abstract
class
OffsetWindowFunctionMeta[INPUT <: OffsetWindowFunction] extends ExprMeta[INPUT]
Spark 3.2-specific replacement for com.nvidia.spark.rapids.OffsetWindowFunctionMeta.
Spark 3.2-specific replacement for com.nvidia.spark.rapids.OffsetWindowFunctionMeta. This is required primarily for two reasons:
- com.nvidia.spark.rapids.OffsetWindowFunctionMeta (compiled against Spark 3.0.x)
fails class load in Spark 3.1.x. (
expr.inputis not recognized as an Expression.) 2. The semantics of offsets in LAG() are reversed/negated in Spark 3.1.1. E.g. The expressionLAG(col, 5)causes Lag.offset to be set to-5, as opposed to5, in prior versions of Spark. 3. IGNORE NULL and RESPECT NULL was added in 3.2.0 and we need to fall back until this can be supported. This class adjusts the LAG offset to use similar semantics to Spark 3.0.x.
- com.nvidia.spark.rapids.OffsetWindowFunctionMeta (compiled against Spark 3.0.x)
fails class load in Spark 3.1.x. (
- trait OrcShims320untilAllBase extends AnyRef
- class RapidsCsvScanMeta extends ScanMeta[CSVScan]
- class RapidsOrcScanMeta extends ScanMeta[OrcScan]
- class RapidsParquetScanMeta extends ScanMeta[ParquetScan]
- trait RebaseShims extends AnyRef
- trait ShimBinaryExecNode extends SparkPlan with BinaryExecNode
- trait ShimBinaryExpression extends BinaryExpression
-
trait
ShimBroadcastExchangeLike extends Exchange with BroadcastExchangeLike
This shim handles the completion future differences between Apache Spark and Databricks.
- trait ShimExpression extends Expression
- abstract class ShimFilePartitionReaderFactory extends FilePartitionReaderFactory
- trait ShimSparkPlan extends SparkPlan
-
trait
ShimSupportsRuntimeFiltering extends SupportsRuntimeFiltering
Shim interface for Apache Spark's SupportsRuntimeFiltering interface which was added in Spark 3.2.0.
- trait ShimTernaryExpression extends TernaryExpression
- trait ShimUnaryCommand extends LogicalPlan with UnaryCommand
- trait ShimUnaryExecNode extends SparkPlan with UnaryExecNode
- trait ShimUnaryExpression extends UnaryExpression
- trait Spark31Xuntil33XShims extends SparkShims
-
trait
Spark320PlusNonDBShims extends SparkShims
Shim methods that can be compiled with every supported 3.2.0+ except Databricks versions
-
trait
Spark320PlusShims extends SparkShims with RebaseShims with Logging
Shim base class that can be compiled with every supported 3.2.0+
- trait Spark320until340Shims extends SparkShims
Value Members
-
object
AQEUtils
Utility methods for manipulating Catalyst classes involved in Adaptive Query Execution
- object AggregationTagging
- object AnsiUtil
- object GlobalLimitShims
- object GpuDataSourceRDD extends Serializable
- object GpuHashPartitioning extends Serializable
-
object
GpuIntervalUtils
Should not support in this Shim
- object GpuParquetCrypto
- object GpuTypeShims
- object GpuWindowUtil
- object HashUtils extends Arm
- object OrcReadingShims
- object OrcShims extends OrcShims320untilAllBase
- object ParquetFieldIdShims
- object ParquetSchemaClipShims
- object ParquetStringPredShims
- object ParquetTimestampNTZShims
- object RapidsFileSourceMetaUtils
- object RegExpShim
- object SparkShimImpl extends Spark320PlusShims with Spark320PlusNonDBShims with Spark31Xuntil33XShims with Spark320until340Shims
-
object
TypeSigUtil extends TypeSigUtilBase
Add DayTimeIntervalType and YearMonthIntervalType support
- object YearParseUtil