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

Optimize JSON writer #13144

Merged
merged 13 commits into from
Apr 24, 2023
+276 −188
Merged

Optimize JSON writer #13144

Show file tree
Hide file tree
Changes from 8 commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
35fb4bd
reduce benchmark runtime by skipping unrequired combinations
karthikeyann Apr 16, 2023
a2c2004
Optimize List and Struct joining methods
karthikeyann Apr 16, 2023
cb24134
update default rows_per_chunk in cython
karthikeyann Apr 16, 2023
d43cad3
cleanup comments
karthikeyann Apr 18, 2023
3d9acc9
Merge branch 'branch-23.06' into enh-json_writer_opt
karthikeyann Apr 18, 2023
d1d6ccb
default value for append_colon
karthikeyann Apr 18, 2023
e88c924
address review comments
karthikeyann Apr 19, 2023
25c9121
Merge branch 'branch-23.06' of github.com:rapidsai/cudf into enh-json…
karthikeyann Apr 19, 2023
64ca53a
address review comments
karthikeyann Apr 20, 2023
14ec55c
lambda to functor
karthikeyann Apr 20, 2023
4deaaa5
Merge branch 'branch-23.06' of github.com:rapidsai/cudf into enh-json…
karthikeyann Apr 20, 2023
9291b4d
address review comments
karthikeyann Apr 21, 2023
92de121
Merge branch 'branch-23.06' of github.com:rapidsai/cudf into enh-json…
karthikeyann Apr 21, 2023
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
20 changes: 13 additions & 7 deletions cpp/benchmarks/io/json/json_writer.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -72,13 +72,24 @@ void BM_json_write_io(nvbench::state& state, nvbench::type_list<nvbench::enum_ty

cuio_source_sink_pair source_sink(source_type);
cudf::io::json_writer_options write_opts =
cudf::io::json_writer_options::builder(source_sink.make_sink_info(), view).na_rep("null");
cudf::io::json_writer_options::builder(source_sink.make_sink_info(), view)
.na_rep("null")
.rows_per_chunk(view.num_rows() / 10);

json_write_common(write_opts, source_sink, data_size, state);
}

void BM_json_writer_options(nvbench::state& state)
{
auto const source_type = io_type::HOST_BUFFER;
bool const json_lines = state.get_int64("json_lines");
bool const include_nulls = state.get_int64("include_nulls");
auto const rows_per_chunk = state.get_int64("rows_per_chunk");

if (!(json_lines == false and include_nulls == false) and rows_per_chunk != 1 << 20) {
karthikeyann marked this conversation as resolved.
Show resolved Hide resolved
state.skip("Skipping for unrequired rows_per_chunk combinations");
return;
}
auto const d_type = get_type_or_group({static_cast<int32_t>(data_type::INTEGRAL),
static_cast<int32_t>(data_type::FLOAT),
static_cast<int32_t>(data_type::DECIMAL),
Expand All @@ -88,11 +99,6 @@ void BM_json_writer_options(nvbench::state& state)
static_cast<int32_t>(data_type::LIST),
static_cast<int32_t>(data_type::STRUCT)});

auto const source_type = io_type::HOST_BUFFER;
bool const json_lines = state.get_int64("json_lines");
bool const include_nulls = state.get_int64("include_nulls");
auto const rows_per_chunk = state.get_int64("rows_per_chunk");

auto const tbl = create_random_table(
cycle_dtypes(d_type, num_cols), table_size_bytes{data_size}, data_profile_builder());
auto const view = tbl->view();
Expand Down Expand Up @@ -122,4 +128,4 @@ NVBENCH_BENCH(BM_json_writer_options)
.set_min_samples(4)
.add_int64_axis("json_lines", {false, true})
.add_int64_axis("include_nulls", {false, true})
.add_int64_power_of_two_axis("rows_per_chunk", nvbench::range(10, 20, 2));
.add_int64_power_of_two_axis("rows_per_chunk", {10, 15, 16, 18, 20});
Loading