datafusion.join.build_input_rows

Build side input rows

Dimensions:None

Available on:Native (1)

Interface Metrics (1)

Native

build_input_rows

Number of rows received from the build side of a join

Dimensions:None

Sources

build_input_rowsgithub.com

Technical Annotations (40)

Configuration Parameters (4)

DATAFUSION_OPTIMIZER_REPARTITION_JOINSrecommended: true

Forces optimizer to consider partitioned joins

DATAFUSION_OPTIMIZER_HASH_JOIN_SINGLE_PARTITION_THRESHOLDrecommended: 0

Disables single partition threshold to force partitioned joins

DATAFUSION_OPTIMIZER_HASH_JOIN_SINGLE_PARTITION_THRESHOLD_ROWSrecommended: 0

Disables row count threshold to force partitioned joins

collect_left_threshold

Memory limit threshold to determine if both join sides can fit in memory for dynamic reordering

Error Signatures (1)

output_rows=0log pattern

CLI Commands (4)

EXPLAIN ANALYZE SELECT * FROM t1 LEFT ANTI JOIN (SELECT * FROM t2 WHERE k <> 1) t2 ON t1.k = t2.kdiagnostic

SELECT * FROM lineitem, orders WHERE l_orderkey = o_orderkey AND o_orderkey = 1 AND l_quantity < (SELECT avg(l_quantity) FROM lineitem WHERE l_orderkey = o_orderkey);

diagnostic

cargo run --profile release-nonlto --bin dfbench tpcds --query 99 --iterations 3 --path benchmarks/data/tpcds_sf1 --query_path datafusion/core/tests/tpc-ds --prefer_hash_join true

diagnostic

datafusion-cli -c "select sum(l_extendedprice) / 7.0 as avg_yearly from lineitem, part where p_partkey = l_partkey and p_brand = 'Brand#23' and p_container = 'MED BOX' and l_quantity < (select 0.2 * avg(l_quantity) from lineitem where l_partkey = p_partkey);"

diagnostic

Technical References (31)

batch_sizeconfiguration parameterCoalesceBatchesExeccomponentBatchSplittercomponentBatchCoalescercomponentSortMergeJoincomponentHashJoincomponentHashJoinExeccomponentLEFT ANTI JOINconceptRIGHT ANTI JOINconceptfast-path optimizationconceptjoin parameterizationconceptpredicate pushdownconceptjoin orderingconceptcardinality estimationconceptTPC-Hconceptbuild sideconceptprobe sideconceptColumnStatisticscomponentTable Statisticscomponentjoin cardinality estimationconceptfilter selectivityconceptCollectLeftcomponentPartitionedcomponentTPC-H query 7conceptTPC-H query 21conceptTPC-H query 4conceptstar schemaconceptright deep treeconceptEXPLAIN ANALYZEcomponentRecordBatchcomponentequi join columnsconcept

Related Insights (10)

Join operators produce non-uniform batch sizes causing memory and performance issueswarning

▸

Hash join with empty build side reports zero output rows despite producing datawarning

▸

Join parameterization missing causes full table scans on selective queriescritical

▸

Suboptimal join ordering on nested TPC-H querieswarning

▸

Hash join input order swap can cause 1000x performance degradationcritical

▸

Join disasters can occur without proper join reordering and statisticscritical

▸

Hash join optimizer selects non-partitioned mode causing 52x slower query executioncritical

▸

Suboptimal hash join build-side selection causes performance degradationwarning

▸

Suboptimal join order causes 60% query performance degradation on multi-table joinswarning

▸

Hash join build side must use smaller input to avoid memory pressure and slow probe phasewarning

▸