Skip to content

Reduce nvbench parameter space for high-multiplicity and sort-merge join benchmarks #21076

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
shrshi wants to merge 1 commit into
base: main
Choose a base branch
from
+12 −10
Open

Reduce nvbench parameter space for high-multiplicity and sort-merge join benchmarks #21076

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
14 changes: 7 additions & 7 deletions cpp/benchmarks/join/multiplicity_join.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,15 +107,15 @@ void nvbench_hm_full_join(nvbench::state& state,

NVBENCH_BENCH_TYPES(nvbench_hm_inner_join,
NVBENCH_TYPE_AXES(JOIN_NULLABLE_RANGE,
JOIN_NULL_EQUALITY,
DEFAULT_JOIN_NULL_EQUALITY,
DEFAULT_JOIN_DATATYPES,
JOIN_ALGORITHM))
.set_name("high_multiplicity_inner_join")
.set_type_axes_names({"Nullable", "NullEquality", "DataType", "Algorithm"})
.add_int64_axis("num_keys", nvbench::range(1, 5, 1))
.add_int64_axis("num_keys", nvbench::range(1, 3, 1))
.add_int64_axis("left_size", JOIN_SIZE_RANGE)
.add_int64_axis("right_size", JOIN_SIZE_RANGE)
.add_int64_axis("multiplicity", {10, 20, 50, 100, 1'000, 10'000, 50'000});
.add_int64_axis("multiplicity", {100, 1'000, 10'000, 50'000});
Copy link
Member

@PointKernel PointKernel Jan 16, 2026

Choose a reason for hiding this comment

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

Suggested change
.add_int64_axis("multiplicity", {100, 1'000, 10'000, 50'000});
.add_int64_axis("multiplicity", {100, 1'000, 10'000});

no big difference between 10'000 and 50'000 IMO. could be nonsense though


NVBENCH_BENCH_TYPES(nvbench_hm_left_join,
NVBENCH_TYPE_AXES(JOIN_NULLABLE_RANGE,
Expand All @@ -124,10 +124,10 @@ NVBENCH_BENCH_TYPES(nvbench_hm_left_join,
JOIN_ALGORITHM))
.set_name("high_multiplicity_left_join")
.set_type_axes_names({"Nullable", "NullEquality", "DataType", "Algorithm"})
.add_int64_axis("num_keys", nvbench::range(1, 5, 1))
.add_int64_axis("num_keys", nvbench::range(1, 3, 1))
.add_int64_axis("left_size", JOIN_SIZE_RANGE)
.add_int64_axis("right_size", JOIN_SIZE_RANGE)
.add_int64_axis("multiplicity", {10, 20, 50, 100, 1'000, 10'000, 50'000});
.add_int64_axis("multiplicity", {100, 1'000, 10'000, 50'000});

NVBENCH_BENCH_TYPES(nvbench_hm_full_join,
NVBENCH_TYPE_AXES(JOIN_NULLABLE_RANGE,
Expand All @@ -136,7 +136,7 @@ NVBENCH_BENCH_TYPES(nvbench_hm_full_join,
JOIN_ALGORITHM))
.set_name("high_multiplicity_full_join")
.set_type_axes_names({"Nullable", "NullEquality", "DataType", "Algorithm"})
.add_int64_axis("num_keys", nvbench::range(1, 5, 1))
.add_int64_axis("num_keys", nvbench::range(1, 3, 1))
.add_int64_axis("left_size", JOIN_SIZE_RANGE)
.add_int64_axis("right_size", JOIN_SIZE_RANGE)
.add_int64_axis("multiplicity", {10, 20, 50, 100, 1'000, 10'000, 50'000});
.add_int64_axis("multiplicity", {100, 1'000, 10'000, 50'000});
8 changes: 5 additions & 3 deletions cpp/benchmarks/join/sort_merge_join.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -106,9 +106,11 @@ void nvbench_sort_merge_inner_join(nvbench::state& state,
}

// Sort-merge inner join with multiple data types and optional key remapping
NVBENCH_BENCH_TYPES(
nvbench_sort_merge_inner_join,
NVBENCH_TYPE_AXES(JOIN_NULLABLE_RANGE, JOIN_NULL_EQUALITY, JOIN_DATATYPES, JOIN_ALGORITHM))
NVBENCH_BENCH_TYPES(nvbench_sort_merge_inner_join,
NVBENCH_TYPE_AXES(JOIN_NULLABLE_RANGE,
JOIN_NULL_EQUALITY,
DEFAULT_JOIN_DATATYPES,
JOIN_ALGORITHM))
.set_name("sort_merge_inner_join")
.set_type_axes_names({"Nullable", "NullEquality", "DataType", "Algorithm"})
.add_int64_axis("num_keys", nvbench::range(1, 3, 1))
Expand Down
Loading