Fix a bug where index joins would fail with the error
driver should never block. Queries will now fail if the index is unable to load within the time specified by the
index-loader-timeoutconfiguration property and the
Fix a bug that could cause
NullPointerExceptionin spatial join with geometry collections on the build side.
Improve performance for queries with
FULL OUTER JOINwhere join keys have the
Improve performance for
Improve performance of repartitioning data between stages. The optimization can be enabled by the
optimized_repartitioningsession property or the
Add spatial join (broadcast and partitioned) support for
task_partitioned_writer_countsession property to allow setting the number of concurrent writers for partitioned (bucketed) writes.
differential_entropy()functions to compute differential entropy.
Remove syntax support for
CURRENT_PATH. The path information was never used by Presto.
Fix a bug that might lead to corruption when writing sorted table in the recoverable grouped execution mode.
Fix ORC stripe skipping when using bloom filter.
Improve the CPU load on coordinator by reducing the cost of serializing
Improve GC pressure from Parquet reader by constraining the maximum column read size. This can be configured by the
parquet_max_read_block_sizesession property or the
Add support for sub-field pruning when reading Parquet files, so that only necessary sub-fields are extracted from struct columns.
Add configuration property
hive.s3-file-system-type=HADOOP_DEFAULTto allow users to switch different Hadoop file system implementations for
s3://addresses. The corresponding Hadoop File system implementation should be specified in
Fix memory leak in file descriptor during shard compaction. The regression was introduced in 0.219.
Add support for auto-resolving query failures with
Add support to perform additional determinism analysis for queries with
Add detailed determinism analysis result to
AggregationNodeto SPI. Connectors can now push down aggregation to table scan.
ProjectNodeto SPI. Connectors can now push down projection to table scan.
Block#getBlockand remove unnecessary