Skip to content

Latest commit

 

History

History
92 lines (85 loc) · 39 KB

File metadata and controls

92 lines (85 loc) · 39 KB

layout: page title: Configuration nav_order: 16

Gluten Velox backend configurations

Key Default Description
spark.gluten.sql.columnar.backend.velox.IOThreads <undefined> The Size of the IO thread pool in the Connector. This thread pool is used for split preloading and DirectBufferedInput. By default, the value is the same as the maximum task slots per Spark executor.
spark.gluten.sql.columnar.backend.velox.SplitPreloadPerDriver 2 The split preload per task
spark.gluten.sql.columnar.backend.velox.abandonPartialAggregationMinPct 90 If partial aggregation aggregationPct greater than this value, partial aggregation may be early abandoned. Note: this option only works when flushable partial aggregation is enabled. Ignored when spark.gluten.sql.columnar.backend.velox.flushablePartialAggregation=false.
spark.gluten.sql.columnar.backend.velox.abandonPartialAggregationMinRows 100000 If partial aggregation input rows number greater than this value, partial aggregation may be early abandoned. Note: this option only works when flushable partial aggregation is enabled. Ignored when spark.gluten.sql.columnar.backend.velox.flushablePartialAggregation=false.
spark.gluten.sql.columnar.backend.velox.asyncTimeoutOnTaskStopping 30000ms Timeout for asynchronous execution when task is being stopped in Velox backend. It's recommended to set to a number larger than network connection timeout that the possible async tasks are relying on.
spark.gluten.sql.columnar.backend.velox.cacheEnabled false Enable Velox cache, default off. It's recommended to enablesoft-affinity as well when enable velox cache.
spark.gluten.sql.columnar.backend.velox.cachePrefetchMinPct 0 Set prefetch cache min pct for velox file scan
spark.gluten.sql.columnar.backend.velox.checkUsageLeak true Enable check memory usage leak.
spark.gluten.sql.columnar.backend.velox.cudf.batchSize 2147483647 Cudf input batch size after shuffle reader
spark.gluten.sql.columnar.backend.velox.cudf.enableTableScan false Enable cudf table scan
spark.gluten.sql.columnar.backend.velox.cudf.enableValidation true Heuristics you can apply to validate a cuDF/GPU plan and only offload when the entire stage can be fully and profitably executed on GPU
spark.gluten.sql.columnar.backend.velox.cudf.memoryPercent 50 The initial percent of GPU memory to allocate for memory resource for one thread.
spark.gluten.sql.columnar.backend.velox.cudf.memoryResource async GPU RMM memory resource.
spark.gluten.sql.columnar.backend.velox.cudf.shuffleMaxPrefetchBytes 1028MB Maximum bytes to prefetch in CPU memory during GPU shuffle read while waiting for GPU available.
spark.gluten.sql.columnar.backend.velox.directorySizeGuess 32KB Deprecated, rename to spark.gluten.sql.columnar.backend.velox.footerEstimatedSize
spark.gluten.sql.columnar.backend.velox.enableTimestampNtzValidation true Enable validation fallback for TimestampNTZ type. When true (default), any plan containing TimestampNTZ will fall back to Spark execution. Set to false during development/testing of TimestampNTZ support to allow native execution.
spark.gluten.sql.columnar.backend.velox.fileHandleCacheEnabled false Disables caching if false. File handle cache should be disabled if files are mutable, i.e. file content may change while file path stays the same.
spark.gluten.sql.columnar.backend.velox.filePreloadThreshold 1MB Set the file preload threshold for velox file scan, refer to Velox's file-preload-threshold
spark.gluten.sql.columnar.backend.velox.floatingPointMode loose Config used to control the tolerance of floating point operations alignment with Spark. When the mode is set to strict, flushing is disabled for sum(float/double)and avg(float/double). When set to loose, flushing will be enabled.
spark.gluten.sql.columnar.backend.velox.flushablePartialAggregation true Enable flushable aggregation. If true, Gluten will try converting regular aggregation into Velox's flushable aggregation when applicable. A flushable aggregation could emit intermediate result at anytime when memory is full / data reduction ratio is low.
spark.gluten.sql.columnar.backend.velox.footerEstimatedSize 32KB Set the footer estimated size for velox file scan, refer to Velox's footer-estimated-size
spark.gluten.sql.columnar.backend.velox.hashProbe.bloomFilterPushdown.maxSize 0b The maximum byte size of Bloom filter that can be generated from hash probe. When set to 0, no Bloom filter will be generated. To achieve optimal performance, this should not be too larger than the CPU cache size on the host.
spark.gluten.sql.columnar.backend.velox.hashProbe.dynamicFilterPushdown.enabled true Whether hash probe can generate any dynamic filter (including Bloom filter) and push down to upstream operators.
spark.gluten.sql.columnar.backend.velox.loadQuantum 256MB Set the load quantum for velox file scan, recommend to use the default value (256MB) for performance consideration. If Velox cache is enabled, it can be 8MB at most.
spark.gluten.sql.columnar.backend.velox.maxCoalescedBytes 64MB Set the max coalesced bytes for velox file scan
spark.gluten.sql.columnar.backend.velox.maxCoalescedDistance 512KB Set the max coalesced distance bytes for velox file scan
spark.gluten.sql.columnar.backend.velox.maxCompiledRegexes 100 Controls maximum number of compiled regular expression patterns per function instance per thread of execution.
spark.gluten.sql.columnar.backend.velox.maxExtendedPartialAggregationMemory <undefined> Set the max extended memory of partial aggregation in bytes. When this option is set to a value greater than 0, it will override spark.gluten.sql.columnar.backend.velox.maxExtendedPartialAggregationMemoryRatio. Note: this option only works when flushable partial aggregation is enabled. Ignored when spark.gluten.sql.columnar.backend.velox.flushablePartialAggregation=false.
spark.gluten.sql.columnar.backend.velox.maxExtendedPartialAggregationMemoryRatio 0.15 Set the max extended memory of partial aggregation as maxExtendedPartialAggregationMemoryRatio of offheap size. Note: this option only works when flushable partial aggregation is enabled. Ignored when spark.gluten.sql.columnar.backend.velox.flushablePartialAggregation=false.
spark.gluten.sql.columnar.backend.velox.maxPartialAggregationMemory <undefined> Set the max memory of partial aggregation in bytes. When this option is set to a value greater than 0, it will override spark.gluten.sql.columnar.backend.velox.maxPartialAggregationMemoryRatio. Note: this option only works when flushable partial aggregation is enabled. Ignored when spark.gluten.sql.columnar.backend.velox.flushablePartialAggregation=false.
spark.gluten.sql.columnar.backend.velox.maxPartialAggregationMemoryRatio 0.1 Set the max memory of partial aggregation as maxPartialAggregationMemoryRatio of offheap size. Note: this option only works when flushable partial aggregation is enabled. Ignored when spark.gluten.sql.columnar.backend.velox.flushablePartialAggregation=false.
spark.gluten.sql.columnar.backend.velox.maxPartitionsPerWritersSession 10000 Maximum number of partitions per a single table writer instance.
spark.gluten.sql.columnar.backend.velox.maxSpillBytes 100G The maximum file size of a query
spark.gluten.sql.columnar.backend.velox.maxSpillFileSize 1GB The maximum size of a single spill file created
spark.gluten.sql.columnar.backend.velox.maxSpillLevel 4 The max allowed spilling level with zero being the initial spilling level
spark.gluten.sql.columnar.backend.velox.maxSpillRunRows 3M The maximum row size of a single spill run
spark.gluten.sql.columnar.backend.velox.maxTargetFileSize 0b The target file size for each output file when writing data. 0 means no limit on target file size, and the actual file size will be determined by other factors such as max partition number and shuffle batch size.
spark.gluten.sql.columnar.backend.velox.memCacheSize 1GB The memory cache size
spark.gluten.sql.columnar.backend.velox.memInitCapacity 8MB The initial memory capacity to reserve for a newly created Velox query memory pool.
spark.gluten.sql.columnar.backend.velox.memoryPoolCapacityTransferAcrossTasks true Whether to allow memory capacity transfer between memory pools from different tasks.
spark.gluten.sql.columnar.backend.velox.memoryUseHugePages false Use explicit huge pages for Velox memory allocation.
spark.gluten.sql.columnar.backend.velox.orc.scan.enabled true Enable velox orc scan. If disabled, vanilla spark orc scan will be used.
spark.gluten.sql.columnar.backend.velox.orcUseColumnNames true Maps table field names to file field names using names, not indices for ORC files.
spark.gluten.sql.columnar.backend.velox.parquetUseColumnNames true Maps table field names to file field names using names, not indices for Parquet files.
spark.gluten.sql.columnar.backend.velox.prefetchRowGroups 1 Set the prefetch row groups for velox file scan
spark.gluten.sql.columnar.backend.velox.queryTraceEnabled false Enable query tracing flag.
spark.gluten.sql.columnar.backend.velox.reclaimMaxWaitMs 3600000ms The max time in ms to wait for memory reclaim.
spark.gluten.sql.columnar.backend.velox.resizeBatches.shuffleInput true If true, combine small columnar batches together before sending to shuffle. The default minimum output batch size is equal to 0.25 * spark.gluten.sql.columnar.maxBatchSize
spark.gluten.sql.columnar.backend.velox.resizeBatches.shuffleInput.minSize <undefined> The minimum batch size for shuffle. If size of an input batch is smaller than the value, it will be combined with other batches before sending to shuffle. Only functions when spark.gluten.sql.columnar.backend.velox.resizeBatches.shuffleInput is set to true. Default value: 0.25 *
spark.gluten.sql.columnar.backend.velox.resizeBatches.shuffleInputOuptut.minSize <undefined> The minimum batch size for shuffle input and output. If size of an input batch is smaller than the value, it will be combined with other batches before sending to shuffle. The same applies for batches output by shuffle read. Only functions when spark.gluten.sql.columnar.backend.velox.resizeBatches.shuffleInput or spark.gluten.sql.columnar.backend.velox.resizeBatches.shuffleOutput is set to true. Default value: 0.25 *
spark.gluten.sql.columnar.backend.velox.resizeBatches.shuffleOutput false If true, combine small columnar batches together right after shuffle read. The default minimum output batch size is equal to 0.25 * spark.gluten.sql.columnar.maxBatchSize
spark.gluten.sql.columnar.backend.velox.showTaskMetricsWhenFinished false Show velox full task metrics when finished.
spark.gluten.sql.columnar.backend.velox.spillFileSystem local The filesystem used to store spill data. local: The local file system. heap-over-local: Write file to JVM heap if having extra heap space. Otherwise write to local file system.
spark.gluten.sql.columnar.backend.velox.spillStrategy auto none: Disable spill on Velox backend; auto: Let Spark memory manager manage Velox's spilling
spark.gluten.sql.columnar.backend.velox.ssdCacheIOThreads 1 The IO threads for cache promoting
spark.gluten.sql.columnar.backend.velox.ssdCachePath /tmp The folder to store the cache files, better on SSD
spark.gluten.sql.columnar.backend.velox.ssdCacheShards 1 The cache shards
spark.gluten.sql.columnar.backend.velox.ssdCacheSize 1GB The SSD cache size, will do memory caching only if this value = 0
spark.gluten.sql.columnar.backend.velox.ssdCheckpointIntervalBytes 0 Checkpoint after every 'checkpointIntervalBytes' for SSD cache. 0 means no checkpointing.
spark.gluten.sql.columnar.backend.velox.ssdChecksumEnabled false If true, checksum write to SSD is enabled.
spark.gluten.sql.columnar.backend.velox.ssdChecksumReadVerificationEnabled false If true, checksum read verification from SSD is enabled.
spark.gluten.sql.columnar.backend.velox.ssdDisableFileCow false True if copy on write should be disabled.
spark.gluten.sql.columnar.backend.velox.ssdODirect false The O_DIRECT flag for cache writing
spark.gluten.sql.columnar.backend.velox.valueStream.dynamicFilter.enabled false Whether to apply dynamic filters pushed down from hash probe in the ValueStream (shuffle reader) operator to filter rows before they reach the hash join.
spark.gluten.sql.enable.enhancedFeatures true Enable some features including iceberg native write and other features.
spark.gluten.sql.rewrite.castArrayToString true When true, rewrite cast(array as String) to concat('[', array_join(array, ', ', null), ']') to allow offloading to Velox.
spark.gluten.velox.broadcast.build.targetBytesPerThread 32MB It is used to calculate the number of hash table build threads. Based on our testing across various thresholds (1MB to 128MB), we recommend a value of 32MB or 64MB, as these consistently provided the most significant performance gains.
spark.gluten.velox.castFromVarcharAddTrimNode false If true, will add a trim node which has the same semantic as vanilla Spark to CAST-from-varchar.Otherwise, do nothing.

Gluten Velox backend experimental configurations

Key Default Description
spark.gluten.velox.abandonDedupHashMap.minPct 0 Experimental: abandon hashmap build if duplicated rows are more than this percentile. Value is integer based and range is [0, 100].
spark.gluten.velox.abandonDedupHashMap.minRows 100000 Experimental: abandon hashmap build if duplicated rows more than this number.
spark.gluten.velox.joinBuildVectorHasherMaxNumDistinct 1000000 Experimental: maximum number of distinct values to keep when merging vector hashers in join HashBuild.
spark.gluten.velox.minTableRowsForParallelJoinBuild 1000 Experimental: the minimum number of table rows that can trigger the parallel hash join table build.
spark.gluten.velox.offHeapBroadcastBuildRelation.enabled false Experimental: If enabled, broadcast build relation will use offheap memory. Otherwise, broadcast build relation will use onheap memory.