Skip to content
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

Fix cudf::hash_join output size for struct joins #9107

Merged
merged 4 commits into from
Aug 25, 2021
Merged

Fix cudf::hash_join output size for struct joins #9107

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
add330b
Fix cudf::hash_join output size for struct joins
jlowe Aug 24, 2021
b4cab63
Update variable names, perform table check in each test case separately
jlowe Aug 24, 2021
8948543
Use structured binding to unpack pair of tables
jlowe Aug 24, 2021
bba3c49
Fix variable names
jlowe Aug 25, 2021
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
31 changes: 22 additions & 9 deletions cpp/src/join/hash_join.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -349,11 +349,15 @@ std::size_t hash_join::hash_join_impl::inner_join_size(cudf::table_view const& p
CUDF_FUNC_RANGE();
CUDF_EXPECTS(_hash_table, "Hash table of hash join is null.");

auto build_table = cudf::table_device_view::create(_build, stream);
auto probe_table = cudf::table_device_view::create(probe, stream);
auto flattened_probe = structs::detail::flatten_nested_columns(
probe, {}, {}, structs::detail::column_nullability::FORCE);
auto const flattened_probe_table = std::get<0>(flattened_probe);

auto build_table = cudf::table_device_view::create(_build, stream);
auto d_flattened_probe_table = cudf::table_device_view::create(flattened_probe_table, stream);
jlowe marked this conversation as resolved.
Show resolved Hide resolved

return cudf::detail::compute_join_output_size<cudf::detail::join_kind::INNER_JOIN>(
*build_table, *probe_table, *_hash_table, compare_nulls, stream);
*build_table, *d_flattened_probe_table, *_hash_table, compare_nulls, stream);
}

std::size_t hash_join::hash_join_impl::left_join_size(cudf::table_view const& probe,
Expand All @@ -365,11 +369,15 @@ std::size_t hash_join::hash_join_impl::left_join_size(cudf::table_view const& pr
// Trivial left join case - exit early
if (!_hash_table) { return probe.num_rows(); }

auto build_table = cudf::table_device_view::create(_build, stream);
auto probe_table = cudf::table_device_view::create(probe, stream);
auto flattened_probe = structs::detail::flatten_nested_columns(
probe, {}, {}, structs::detail::column_nullability::FORCE);
auto const flattened_probe_table = std::get<0>(flattened_probe);

auto build_table = cudf::table_device_view::create(_build, stream);
auto d_flattened_probe_table = cudf::table_device_view::create(flattened_probe_table, stream);
jlowe marked this conversation as resolved.
Show resolved Hide resolved

return cudf::detail::compute_join_output_size<cudf::detail::join_kind::LEFT_JOIN>(
*build_table, *probe_table, *_hash_table, compare_nulls, stream);
*build_table, *d_flattened_probe_table, *_hash_table, compare_nulls, stream);
}

std::size_t hash_join::hash_join_impl::full_join_size(cudf::table_view const& probe,
Expand All @@ -382,10 +390,15 @@ std::size_t hash_join::hash_join_impl::full_join_size(cudf::table_view const& pr
// Trivial left join case - exit early
if (!_hash_table) { return probe.num_rows(); }

auto build_table = cudf::table_device_view::create(_build, stream);
auto probe_table = cudf::table_device_view::create(probe, stream);
auto flattened_probe = structs::detail::flatten_nested_columns(
probe, {}, {}, structs::detail::column_nullability::FORCE);
auto const flattened_probe_table = std::get<0>(flattened_probe);

auto build_table = cudf::table_device_view::create(_build, stream);
auto d_flattened_probe_table = cudf::table_device_view::create(flattened_probe_table, stream);
jlowe marked this conversation as resolved.
Show resolved Hide resolved

return get_full_join_size(*build_table, *probe_table, *_hash_table, compare_nulls, stream, mr);
return get_full_join_size(
*build_table, *d_flattened_probe_table, *_hash_table, compare_nulls, stream, mr);
}

template <cudf::detail::join_kind JoinKind>
Expand Down
136 changes: 79 additions & 57 deletions cpp/tests/join/join_tests.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -44,6 +44,28 @@ constexpr cudf::size_type NoneValue =
std::numeric_limits<cudf::size_type>::min(); // TODO: how to test if this isn't public?

struct JoinTest : public cudf::test::BaseFixture {
void check_gather_maps(
jlowe marked this conversation as resolved.
Show resolved Hide resolved
cudf::column_view const& left_map,
cudf::column_view const& right_map,
std::pair<std::unique_ptr<rmm::device_uvector<cudf::size_type>>,
std::unique_ptr<rmm::device_uvector<cudf::size_type>>> const& result)
{
auto result_table =
cudf::table_view({cudf::column_view{cudf::data_type{cudf::type_id::INT32},
static_cast<cudf::size_type>(result.first->size()),
result.first->data()},
cudf::column_view{cudf::data_type{cudf::type_id::INT32},
static_cast<cudf::size_type>(result.second->size()),
result.second->data()}});
auto result_sort_order = cudf::sorted_order(result_table);
auto sorted_result = cudf::gather(result_table, *result_sort_order);

cudf::table_view gold({left_map, right_map});
auto gold_sort_order = cudf::sorted_order(gold);
auto sorted_gold = cudf::gather(gold, *gold_sort_order);

CUDF_TEST_EXPECT_TABLES_EQUIVALENT(*sorted_gold, *sorted_result);
}
};

TEST_F(JoinTest, EmptySentinelRepro)
Expand Down Expand Up @@ -1232,28 +1254,9 @@ TEST_F(JoinTest, HashJoinSequentialProbes)
EXPECT_EQ(output_size, size_gold);

auto result = hash_join.full_join(t0, cudf::null_equality::EQUAL, optional_size);
auto result_table =
cudf::table_view({cudf::column_view{cudf::data_type{cudf::type_id::INT32},
static_cast<cudf::size_type>(result.first->size()),
result.first->data()},
cudf::column_view{cudf::data_type{cudf::type_id::INT32},
static_cast<cudf::size_type>(result.second->size()),
result.second->data()}});
auto result_sort_order = cudf::sorted_order(result_table);
auto sorted_result = cudf::gather(result_table, *result_sort_order);

column_wrapper<int32_t> col_gold_0{{NoneValue, NoneValue, NoneValue, NoneValue, 4, 0, 1, 2, 3}};
column_wrapper<int32_t> col_gold_1{{0, 1, 2, 3, 4, NoneValue, NoneValue, NoneValue, NoneValue}};

CVector cols_gold;
cols_gold.push_back(col_gold_0.release());
cols_gold.push_back(col_gold_1.release());

Table gold(std::move(cols_gold));
auto gold_sort_order = cudf::sorted_order(gold.view());
auto sorted_gold = cudf::gather(gold.view(), *gold_sort_order);

CUDF_TEST_EXPECT_TABLES_EQUIVALENT(*sorted_gold, *sorted_result);
check_gather_maps(col_gold_0, col_gold_1, result);
}

{
Expand All @@ -1270,28 +1273,9 @@ TEST_F(JoinTest, HashJoinSequentialProbes)
EXPECT_EQ(output_size, size_gold);

auto result = hash_join.left_join(t0, cudf::null_equality::EQUAL, optional_size);
auto result_table =
cudf::table_view({cudf::column_view{cudf::data_type{cudf::type_id::INT32},
static_cast<cudf::size_type>(result.first->size()),
result.first->data()},
cudf::column_view{cudf::data_type{cudf::type_id::INT32},
static_cast<cudf::size_type>(result.second->size()),
result.second->data()}});
auto result_sort_order = cudf::sorted_order(result_table);
auto sorted_result = cudf::gather(result_table, *result_sort_order);

column_wrapper<int32_t> col_gold_0{{0, 1, 2, 3, 4}};
column_wrapper<int32_t> col_gold_1{{NoneValue, NoneValue, NoneValue, NoneValue, 4}};

CVector cols_gold;
cols_gold.push_back(col_gold_0.release());
cols_gold.push_back(col_gold_1.release());

Table gold(std::move(cols_gold));
auto gold_sort_order = cudf::sorted_order(gold.view());
auto sorted_gold = cudf::gather(gold.view(), *gold_sort_order);

CUDF_TEST_EXPECT_TABLES_EQUIVALENT(*sorted_gold, *sorted_result);
check_gather_maps(col_gold_0, col_gold_1, result);
}

{
Expand All @@ -1308,28 +1292,66 @@ TEST_F(JoinTest, HashJoinSequentialProbes)
EXPECT_EQ(output_size, size_gold);

auto result = hash_join.inner_join(t0, cudf::null_equality::EQUAL, optional_size);
auto result_table =
cudf::table_view({cudf::column_view{cudf::data_type{cudf::type_id::INT32},
static_cast<cudf::size_type>(result.first->size()),
result.first->data()},
cudf::column_view{cudf::data_type{cudf::type_id::INT32},
static_cast<cudf::size_type>(result.second->size()),
result.second->data()}});
auto result_sort_order = cudf::sorted_order(result_table);
auto sorted_result = cudf::gather(result_table, *result_sort_order);

column_wrapper<int32_t> col_gold_0{{2, 4, 0}};
column_wrapper<int32_t> col_gold_1{{1, 1, 4}};
check_gather_maps(col_gold_0, col_gold_1, result);
}
}

TEST_F(JoinTest, HashJoinWithStructsAndNulls)
{
auto col0_names_col = strcol_wrapper{
"Samuel Vimes", "Carrot Ironfoundersson", "Detritus", "Samuel Vimes", "Angua von Überwald"};
auto col0_ages_col = column_wrapper<int32_t>{{48, 27, 351, 31, 25}};

CVector cols_gold;
cols_gold.push_back(col_gold_0.release());
cols_gold.push_back(col_gold_1.release());
auto col0_is_human_col = column_wrapper<bool>{{true, true, false, false, false}, {1, 1, 0, 1, 0}};

Table gold(std::move(cols_gold));
auto gold_sort_order = cudf::sorted_order(gold.view());
auto sorted_gold = cudf::gather(gold.view(), *gold_sort_order);
auto col0 =
cudf::test::structs_column_wrapper{{col0_names_col, col0_ages_col, col0_is_human_col}};
jlowe marked this conversation as resolved.
Show resolved Hide resolved

CUDF_TEST_EXPECT_TABLES_EQUIVALENT(*sorted_gold, *sorted_result);
auto col1_names_col = strcol_wrapper{
"Samuel Vimes", "Detritus", "Detritus", "Carrot Ironfoundersson", "Angua von Überwald"};
auto col1_ages_col = column_wrapper<int32_t>{{48, 35, 351, 22, 25}};

auto col1_is_human_col = column_wrapper<bool>{{true, true, false, false, true}, {1, 1, 0, 1, 1}};

auto col1 =
cudf::test::structs_column_wrapper{{col1_names_col, col1_ages_col, col1_is_human_col}};
jlowe marked this conversation as resolved.
Show resolved Hide resolved

CVector cols0, cols1;
cols0.push_back(col0.release());
cols1.push_back(col1.release());

Table t0(std::move(cols0));
Table t1(std::move(cols1));

auto hash_join = cudf::hash_join(t1, cudf::null_equality::EQUAL);

{
auto output_size = hash_join.left_join_size(t0);
EXPECT_EQ(5, output_size);
auto result = hash_join.left_join(t0, cudf::null_equality::EQUAL, output_size);
column_wrapper<int32_t> col_gold_0{{0, 1, 2, 3, 4}};
column_wrapper<int32_t> col_gold_1{{0, NoneValue, 2, NoneValue, NoneValue}};
check_gather_maps(col_gold_0, col_gold_1, result);
}

{
auto output_size = hash_join.inner_join_size(t0);
EXPECT_EQ(2, output_size);
auto result = hash_join.inner_join(t0, cudf::null_equality::EQUAL, output_size);
column_wrapper<int32_t> col_gold_0{{0, 2}};
column_wrapper<int32_t> col_gold_1{{0, 2}};
check_gather_maps(col_gold_0, col_gold_1, result);
}

{
auto output_size = hash_join.full_join_size(t0);
EXPECT_EQ(8, output_size);
auto result = hash_join.full_join(t0, cudf::null_equality::EQUAL, output_size);
column_wrapper<int32_t> col_gold_0{{NoneValue, NoneValue, NoneValue, 0, 1, 2, 3, 4}};
column_wrapper<int32_t> col_gold_1{{1, 3, 4, 0, NoneValue, 2, NoneValue, NoneValue}};
check_gather_maps(col_gold_0, col_gold_1, result);
}
}

Expand Down