Planner — TPC-H — TPCH-Q03

Engine Compare

Accuracy chart, rows processed ?
	Scan Scan	Seek Seek	Join Probe Join	Sort Sort	Hash Build Hash	Aggregate Agg	Distribute Dist
ClickHouse	Estimation Error Est Err Row Operations Ops 3,999,030 4M Rank		Estimation Error Est Err Row Operations Ops 3,393,107 3.4M Rank	Estimation Error Est Err Row Operations Ops 11,281 11K Rank	Estimation Error Est Err Row Operations Ops 757,254 757K Rank	Estimation Error Est Err Row Operations Ops 29,857 30K Rank	Estimation Error Est Err Row Operations Ops 0 Rank
Databricks	Estimation Error Est Err Row Operations Ops 3,997,498 4M Rank		Estimation Error Est Err Row Operations Ops 975,174 975K Rank	Estimation Error Est Err Row Operations Ops 11,281 11K Rank	Estimation Error Est Err Row Operations Ops 870,989 871K Rank	Estimation Error Est Err Row Operations Ops 3,264,338 3.3M Rank	Estimation Error Est Err Row Operations Ops 0 Rank
DuckDB	Estimation Error Est Err Row Operations Ops 809,889 810K Rank		Estimation Error Est Err Row Operations Ops 779,940 780K Rank	Estimation Error Est Err Row Operations Ops 0 0 Rank	Estimation Error Est Err Row Operations Ops 290,432 290K Rank	Estimation Error Est Err Row Operations Ops 29,857 30K Rank	Estimation Error Est Err Row Operations Ops 0 Rank
PostgreSQL	Estimation Error Est Err Row Operations Ops 902,470 902K Rank	Estimation Error Est Err Row Operations Ops 145,216 145K Rank	Estimation Error Est Err Row Operations Ops 872,521 873K Rank	Estimation Error Est Err Row Operations Ops 29,866 30K Rank	Estimation Error Est Err Row Operations Ops 96,810 97K Rank	Estimation Error Est Err Row Operations Ops 29,856 30K Rank	Estimation Error Est Err Row Operations Ops 0 Rank
SQL Server	Estimation Error Est Err Row Operations Ops 205,022 205K Rank		Estimation Error Est Err Row Operations Ops 156,497 156K Rank	Estimation Error Est Err Row Operations Ops 10 10 Rank	Estimation Error Est Err Row Operations Ops 290,432 290K Rank	Estimation Error Est Err Row Operations Ops 29,857 30K Rank	Estimation Error Est Err Row Operations Ops 0 Rank
Trino	Estimation Error Est Err Row Operations Ops 7,651,215 7.7M Rank		Estimation Error Est Err Row Operations Ops 290,432 290K Rank	Estimation Error Est Err Row Operations Ops 45,124 45K Rank	Estimation Error Est Err Row Operations Ops 3,969,081 4M Rank	Estimation Error Est Err Row Operations Ops 29,857 30K Rank	Estimation Error Est Err Row Operations Ops 0 Rank

Actual Query Plans

Query Plan per Engine ?
	Query Plan
DuckDB	Estimate Actual Operator 10 10 LIMIT 10 10 ∞ SORT SUM(l_extendedprice * (1 - l_discount)), o_orderdate 63865 11281 PROJECT l_orderkey, revenue, o_orderdate, o_shippriority 63865 11281 AGGREGATE SUM(#3) GROUP BY HASH #0, #1, #2 63866 29857 PROJECT l_orderkey, o_orderdate, o_shippriority, l_extendedprice * (1.00 - l_discount) 63866 29857 INNER JOIN HASH ON l_orderkey = o_orderkey 79155 145216 │└INNER JOIN HASH ON o_custkey = c_custkey 37500 29949 │ │└FILTER c_custkey <= 149999 37500 29949 │ │ TABLE SCAN customer WHERE c_mktsegment = 'MACHINERY' 300000 727292 │ TABLE SCAN orders WHERE o_orderdate < '1995-03-15' 1200243 52648 TABLE SCAN lineitem WHERE l_shipdate > '1995-03-15'
Databricks	Estimate Actual Operator 10 10 LIMIT 10 1830000 11281 SORT revenue DESC NULLS LAST, o_orderdate ASC NULLS FIRST 1830000 11281 AGGREGATE SUM(l_extendedprice * (1BD - l_discount)) GROUP BY HASH l_orderkey, o_orderdate, o_shippriority 1830000 11281 DISTRIBUTE HASH ON l_orderkey, o_orderdate, o_shippriority 1830000 11281 AGGREGATE SUM(l_extendedprice * (1BD - l_discount)) GROUP BY HASH l_orderkey, o_orderdate, o_shippriority 1830000 11281 INNER JOIN HASH ON o_orderkey = l_orderkey 1830000 145216 │└DISTRIBUTE GATHER 451000 145216 │ INNER JOIN HASH ON o_custkey = c_custkey 150000 725773 │ │└TABLE SCAN orders WHERE o_orderdate < DATE'1995-03-15' 451000 29949 │ DISTRIBUTE GATHER 6000000 29949 │ TABLE SCAN customer WHERE c_mktsegment = 'MACHINERY'collate UTF8_BINARY 3230000 829958 AGGREGATE SUM(l_extendedprice * (1BD - l_discount)) GROUP BY HASH l_orderkey 1500000 3241776 TABLE SCAN lineitem WHERE l_shipdate > DATE'1995-03-15'
ClickHouse	Estimate Actual Operator - 10 PROJECT l_orderkey, a1 AS revenue, o_orderdate, o_shippriority - 10 LIMIT 10 - 11281 SORT a1, o_orderdate - 11281 PROJECT a1, o_orderdate, l_orderkey, o_shippriority - 11281 AGGREGATE SUM(a5) AS a1 GROUP BY HASH l_orderkey, o_orderdate, o_shippriority - 29857 PROJECT l_orderkey, o_orderdate, o_shippriority, l_extendedprice * (1 - l_discount) AS a5 - 29857 PROJECT l_orderkey, l_extendedprice, l_discount, o_orderdate, o_shippriority - 29857 INNER JOIN HASH ON PROJECTION_75.o_custkey = PROJECTION_72.c_custkey - 29949 │└PROJECT c_custkey - 29949 │ PROJECT c_custkey - 29949 │ TABLE SCAN customer WHERE c_mktsegment = 'MACHINERY' - 151331 PROJECT o_custkey, l_orderkey, l_extendedprice, l_discount, o_orderdate, o_shippriority - 151331 PROJECT l_orderkey, l_extendedprice, l_discount, o_orderdate, o_shippriority, o_custkey - 151331 INNER JOIN HASH ON PROJECTION_81.l_orderkey = PROJECTION_78.o_orderkey - 727305 │└PROJECT o_orderkey, o_orderdate, o_shippriority, o_custkey - 727305 │ PROJECT o_orderkey, o_orderdate, o_shippriority, o_custkey - 727305 │ TABLE SCAN orders WHERE o_orderdate < '1995-03-15' - 3241776 PROJECT l_orderkey, l_extendedprice, l_discount - 3241776 PROJECT l_orderkey, l_extendedprice, l_discount - 3241776 TABLE SCAN lineitem WHERE l_shipdate > '1995-03-15'
PostgreSQL	Estimate Actual Operator 10 10 LIMIT 10 318752 10 SORT SUM(l_extendedprice * ('1' - l_discount)), o_orderdate 318752 11281 AGGREGATE SUM(l_extendedprice * ('1' - l_discount)) GROUP BY SORT l_orderkey, o_orderdate, o_shippriority 398439 29856 SORT l_orderkey, o_orderdate, o_shippriority 398439 29856 INNER JOIN LOOP ON l_orderkey = o_orderkey 61994 48405 │└INNER JOIN HASH ON o_custkey = c_custkey 38268 29949 │ │└TABLE SCAN customer AS customer WHERE c_mktsegment = 'MACHINERY' 911229 727305 │ TABLE SCAN orders AS orders WHERE o_orderdate < '1995-03-15' 1016512 145216 TABLE SEEK lineitem AS lineitem WHERE l_shipdate > '1995-03-15'
Trino	Estimate Actual Operator 10 10 PROJECT l_orderkey, sum AS revenue, o_orderdate, o_shippriority 10 10 LIMIT 10 10 11281 SORT sum, o_orderdate 10 11281 DISTRIBUTE GATHER 10 11281 SORT sum, o_orderdate 10 11281 DISTRIBUTE GATHER 10 11281 SORT sum, o_orderdate 10 11281 DISTRIBUTE ROUND ROBIN 10 11281 SORT sum, o_orderdate 470322 11281 AGGREGATE SUM(expr) AS sum GROUP BY HASH l_orderkey, o_orderdate, o_shippriority 470322 29857 PROJECT l_extendedprice * (1.0 - l_discount) AS expr, l_orderkey, o_orderdate, o_shippriority 470322 29857 INNER JOIN HASH ON l_orderkey = o_orderkey 6001215 3241776 │└FILTER l_shipdate > '1995-03-15' 6001215 6001215 │ TABLE SCAN lineitem 218740 145216 DISTRIBUTE HASH ON o_orderkey 218740 145216 INNER JOIN HASH ON o_custkey = c_custkey 1500000 727305 │└FILTER o_orderdate < '1995-03-15' 1500000 1500000 │ TABLE SCAN orders 30000 29949 DISTRIBUTE GATHER 150000 29949 FILTER c_mktsegment = 'MACHINERY' 150000 150000 TABLE SCAN customer
SQL Server	Estimate Actual Operator 10 10 LIMIT 0 10 10 SORT Expr1009, o_orderdate 222132 11281 INNER JOIN HASH ON l_orderkey = o_orderkey 228150 145216 │└INNER JOIN HASH ON o_custkey = c_custkey 29868 29949 │ │└TABLE SCAN customer WHERE c_mktsegment = 'MACHINERY' 72698 145216 │ TABLE SCAN orders WHERE o_orderdate < '1995-03-15' AND BLOOM(o_custkey) 124448 11281 AGGREGATE SUM(Expr1010) AS Expr1009 GROUP BY HASH l_orderkey 323565 29857 PROJECT l_extendedprice * (1. - l_discount) AS Expr1010 323565 29857 TABLE SCAN lineitem WHERE l_shipdate > '1995-03-15' AND BLOOM(l_orderkey)

Commentary

Optimal Join Order

Optimisers only seeking left deep trees will consume signficant memory while executing the query.

The key optimisation in join ordering is to create a bushy join on customer and orders before joining to lineitem. This allows both l_shipdate > '1995-03-15' and o_orderdate < '1995-03-15' to be be pre-filtered before the join to lineitem greatly reducing the amount of data to be processed.

Optimisers who understand bloom filters will be able to harvest further filters by combining the customer and orders filters before accessing lineitem.

Heap Sort

The LIMIT 10 in the query is best optimised with heap sort nodes. This is typically done with a special sort operator that can cut the stream early. Good optimisers and execution engines find this optimisation and reduce the amount of data to be sorted before emitting the top 10.

In the dbprove tool, this shows up a the sort not only emitting a few rows.