Skip to content

chore: Add existence (semi / anti ) benchmarks for hashjoinexec #21821

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
coderfender wants to merge 4 commits into
base: main
Choose a base branch
from
+531 −0
Open

chore: Add existence (semi / anti ) benchmarks for hashjoinexec #21821

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
a9d6d68
add_existence_benchmarks_hashjoin
coderfender Apr 24, 2026
6eb40b7
add_existence_benchmarks_hashjoin
coderfender Apr 24, 2026
57cfc3e
Merge branch 'main' into implement_additional_benchmarks_hashjoin
coderfender Apr 24, 2026
ad83914
Merge branch 'main' into implement_additional_benchmarks_hashjoin
coderfender Apr 25, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
192 changes: 192 additions & 0 deletions benchmarks/src/hj.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -303,6 +303,198 @@ const HASH_QUERIES: &[HashJoinQuery] = &[
build_size: "100K_(20%_dups)",
probe_size: "60M",
},
// RightSemi Join benchmarks with Int32 keys
// Q16: RightSemi, 100% Density, 100% Hit rate
HashJoinQuery {
sql: r###"SELECT l.k
Copy link
Copy Markdown
Contributor

@2010YOUY01 2010YOUY01 Apr 25, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It might be clearer to express these directly using RIGHT SEMI JOIN, for example:

DataFusion CLI v53.1.0
> select count(*)
from generate_series(100) as t1(v1)
right semi join generate_series(100000) as t2(v1)
on t1.v1 > t2.v1;
+----------+
| count(*) |
+----------+
| 100      |
+----------+
1 row(s) fetched.
Elapsed 0.077 seconds.

> select count(*)
from generate_series(100) as t1(v1)
right anti join generate_series(100000) as t2(v1)
on t1.v1 > t2.v1;
+----------+
| count(*) |
+----------+
| 99901    |
+----------+
1 row(s) fetched.
Elapsed 0.007 seconds.

Though, I'm not sure if it's standard SQL 🤔 , but df have them and it's easier to read.

FROM (
SELECT CAST(l_suppkey AS INT) as k FROM lineitem
) l
WHERE EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 1.0,
prob_hit: 1.0,
build_size: "100K",
probe_size: "60M_RightSemi",
},
// Q17: RightSemi, 100% Density, 10% Hit rate
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(CASE WHEN l_suppkey % 10 = 0 THEN l_suppkey ELSE l_suppkey + 1000000 END AS INT) as k
FROM lineitem
) l
WHERE EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 1.0,
prob_hit: 0.1,
build_size: "100K",
probe_size: "60M_RightSemi",
},
// Q18: RightSemi, 50% Density, 100% Hit rate
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(l_suppkey * 2 AS INT) as k FROM lineitem
) l
WHERE EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey * 2 AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 0.5,
prob_hit: 1.0,
build_size: "100K",
probe_size: "60M_RightSemi",
},
// Q19: RightSemi, 50% Density, 10% Hit rate
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(CASE
WHEN l_suppkey % 10 = 0 THEN l_suppkey * 2
WHEN l_suppkey % 10 < 9 THEN l_suppkey * 2 + 1
ELSE l_suppkey * 2 + 1000000
END AS INT) as k
FROM lineitem
) l
WHERE EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey * 2 AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 0.5,
prob_hit: 0.1,
build_size: "100K",
probe_size: "60M_RightSemi",
},
// Q20: RightSemi, 10% Density, 100% Hit rate
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(l_suppkey * 10 AS INT) as k FROM lineitem
) l
WHERE EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey * 10 AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 0.1,
prob_hit: 1.0,
build_size: "100K",
probe_size: "60M_RightSemi",
},
// Q21: RightSemi, 10% Density, 10% Hit rate
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(CASE
WHEN l_suppkey % 10 = 0 THEN l_suppkey * 10
WHEN l_suppkey % 10 < 9 THEN l_suppkey * 10 + 1
ELSE l_suppkey * 10 + 1000000
END AS INT) as k
FROM lineitem
) l
WHERE EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey * 10 AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 0.1,
prob_hit: 0.1,
build_size: "100K",
probe_size: "60M_RightSemi",
},
// RightAnti Join benchmarks with Int32 keys
// Q22: RightAnti, 100% Density, 100% Hit rate (no output)
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(l_suppkey AS INT) as k FROM lineitem
) l
WHERE NOT EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 1.0,
prob_hit: 1.0,
build_size: "100K",
probe_size: "60M_RightAnti",
},
// Q23: RightAnti, 100% Density, 10% Hit rate (90% output)
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(CASE WHEN l_suppkey % 10 = 0 THEN l_suppkey ELSE l_suppkey + 1000000 END AS INT) as k
FROM lineitem
) l
WHERE NOT EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 1.0,
prob_hit: 0.1,
build_size: "100K",
probe_size: "60M_RightAnti",
},
// Q24: RightAnti, 50% Density, 100% Hit rate (no output)
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(l_suppkey * 2 AS INT) as k FROM lineitem
) l
WHERE NOT EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey * 2 AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 0.5,
prob_hit: 1.0,
build_size: "100K",
probe_size: "60M_RightAnti",
},
// Q25: RightAnti, 50% Density, 10% Hit rate (90% output)
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(CASE
WHEN l_suppkey % 10 = 0 THEN l_suppkey * 2
WHEN l_suppkey % 10 < 9 THEN l_suppkey * 2 + 1
ELSE l_suppkey * 2 + 1000000
END AS INT) as k
FROM lineitem
) l
WHERE NOT EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey * 2 AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 0.5,
prob_hit: 0.1,
build_size: "100K",
probe_size: "60M_RightAnti",
},
// Q26: RightAnti, 10% Density, 100% Hit rate (no output)
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(l_suppkey * 10 AS INT) as k FROM lineitem
) l
WHERE NOT EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey * 10 AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 0.1,
prob_hit: 1.0,
build_size: "100K",
probe_size: "60M_RightAnti",
},
// Q27: RightAnti, 10% Density, 10% Hit rate (90% output)
HashJoinQuery {
sql: r###"SELECT l.k
FROM (
SELECT CAST(CASE
WHEN l_suppkey % 10 = 0 THEN l_suppkey * 10
WHEN l_suppkey % 10 < 9 THEN l_suppkey * 10 + 1
ELSE l_suppkey * 10 + 1000000
END AS INT) as k
FROM lineitem
) l
WHERE NOT EXISTS (
SELECT 1 FROM (SELECT CAST(s_suppkey * 10 AS INT) as k FROM supplier) s WHERE s.k = l.k
)"###,
density: 0.1,
prob_hit: 0.1,
build_size: "100K",
probe_size: "60M_RightAnti",
},
];

impl RunOpt {
Expand Down
5 changes: 5 additions & 0 deletions datafusion/physical-plan/Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -108,3 +108,8 @@ required-features = ["test_utils"]
harness = false
name = "aggregate_vectorized"
required-features = ["test_utils"]

[[bench]]
harness = false
name = "hash_join_semi_anti"
required-features = ["test_utils"]
Loading
Loading