Skip to content

+Perfect GroupBy and some old GroupBy algos changes #35

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
wants to merge 14 commits into
base: main
Choose a base branch
from
Open

+Perfect GroupBy and some old GroupBy algos changes #35

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
e5aa7ba
renamed global groupby to GroupByGLobal
bagrorg Jul 12, 2022
294e559
+ groupby abstraction:
bagrorg Jul 12, 2022
1e4e89f
+ groupby inheritance for local algorithm
bagrorg Jul 12, 2022
bb6597f
+ size for groupby
bagrorg Jul 12, 2022
8bd88ba
removed useless sizeof
bagrorg Jul 12, 2022
59c8ae9
+ perfect group by
bagrorg Jul 12, 2022
5c1fd9a
On branch groupby_branch
bagrorg Jul 12, 2022
c6964ae
executors -> threads_count and better local groupby for loop
bagrorg Jul 12, 2022
b2ac6ba
+ options for perfect groupby
bagrorg Jul 12, 2022
b0e3562
+ formatting
bagrorg Jul 12, 2022
69b3092
+ formatting on options.hpp
bagrorg Jul 12, 2022
7111ef6
Merge branch 'main' into groupby_branch
bagrorg Aug 19, 2022
9e35746
fixed clean segfault and formatting
bagrorg Aug 19, 2022
8020042
+ tests for new groupby algos
bagrorg Aug 23, 2022
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
3 changes: 2 additions & 1 deletion groupby/perfect_groupby.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,7 +91,8 @@ void PerfectGroupBy::_run(const size_t buf_size, Meter &meter) {
h.single_task<class clean>([=]() {
for (size_t i = 0; i < groups_count * threads_count; i++) {
ht_v[i] = 0;
o[i] = 0;
if (i < groups_count)
o[i] = 0;
Copy link
Owner

Choose a reason for hiding this comment

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

What's the logic behind this one?

Copy link
Collaborator Author

Choose a reason for hiding this comment

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

Here we want to clear all output from previous iteration using single_task
However, groups_count * threads_count may be bigger than output array size, so we need to control it to prevent SEGFAULT
This is not so clear, I wanted to do it in one for loop, but is it better to separate this two operations into two single_tasks?

Copy link
Owner

Choose a reason for hiding this comment

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

I guess no big difference since it's a single task

}
});
}).wait();
Expand Down
3 changes: 2 additions & 1 deletion tests/dwarf_tests/dwarf_tests.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,7 +70,8 @@ GENERATE_TEST_SUITE(ReduceDPCPP);
GENERATE_TEST_SUITE(HashBuild);
GENERATE_TEST_SUITE(NestedLoopJoin);
GENERATE_TEST_SUITE(CuckooHashBuild);
GENERATE_TEST_SUITE_GROUPBY(GroupBy);
GENERATE_TEST_SUITE_GROUPBY(GroupByGlobal);
GENERATE_TEST_SUITE_GROUPBY(PerfectGroupBy);
GENERATE_TEST_SUITE_GROUPBY(GroupByLocal);
GENERATE_TEST_SUITE(Join);
GENERATE_TEST_SUITE(HashBuildNonBitmask);
Expand Down
4 changes: 3 additions & 1 deletion tests/dwarf_tests/dwarfs.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,9 @@

#include "constant/constant.hpp"
#include "groupby/groupby.hpp"
#include "groupby/groupby_global.hpp"
#include "groupby/groupby_local.hpp"
#include "groupby/perfect_groupby.hpp"
#include "hash/cuckoo_hash_build.hpp"
#include "hash/hash_build.hpp"
#include "hash/hash_build_non_bitmask.hpp"
Expand All @@ -15,4 +17,4 @@
#include "reduce/reduce.hpp"
#include "scan/scan.hpp"
#include "sort/radix.hpp"
#include "sort/tbbsort.hpp"
#include "sort/tbbsort.hpp"
8 changes: 4 additions & 4 deletions tests/dwarf_tests/utils.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,13 +37,13 @@ std::unique_ptr<RunOptions> get_cpu_test_opts(size_t size) {
}

std::unique_ptr<RunOptions> get_cpu_test_opts_groupby(size_t size) {
return std::make_unique<GroupByRunOptions>(*get_cpu_test_opts(size), 64,
1024);
return std::make_unique<GroupByRunOptions>(*get_cpu_test_opts(size), 64, 1024,
32);
}

std::unique_ptr<RunOptions> get_gpu_test_opts_groupby(size_t size) {
return std::make_unique<GroupByRunOptions>(*get_gpu_test_opts(size), 64,
1024);
return std::make_unique<GroupByRunOptions>(*get_gpu_test_opts(size), 64, 1024,
32);
}

std::string get_kernels_root_tests() {
Expand Down