database/doris
2
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

[improvement](bitmap) Using set to store a small number of elements to improve performance (#19973)

Browse Source 
Test on SSB 100g:

select lo_suppkey, count(distinct lo_linenumber) from lineorder group by lo_suppkey;
exec time: 4.388s

create materialized view:

create materialized view customer_uv as select lo_suppkey, bitmap_union(to_bitmap(lo_linenumber)) from lineorder group by lo_suppkey;
select lo_suppkey, count(distinct lo_linenumber) from lineorder group by lo_suppkey;
exec time: 12.908s

test with the patch, exec time: 5.790s
This commit is contained in:
Jerry Hu
2023-05-31 16:13:42 +08:00
committed by GitHub
parent b53c42636e
commit c03a19ea23
8 changed files with 689 additions and 96 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
be/src/vec/functions/function_bitmap.cpp
View File

@ -127,14 +127,12 @@ struct ToBitmap {
size_t size = col->size();
for (size_t i = 0; i < size; ++i) {
if (arg_is_nullable && ((*nullmap)[i])) {
continue;
} else {
int64_t int_value = col->get_data()[i];
if (LIKELY(int_value >= 0)) {
res_data[i].add(int_value);
if constexpr (arg_is_nullable) {
if ((*nullmap)[i]) {
continue;
}
}
res_data[i].add(col->get_data()[i]);
}
}
}