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

bloom filter for ngramms #12893

Merged
merged 3 commits into from
Dec 25, 2024
Merged

bloom filter for ngramms #12893

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
0ca7ce6
bloom filter for ngramms
ivanmorozov333 Dec 23, 2024
5709a70
fixes and improvements
ivanmorozov333 Dec 24, 2024
45189a5
improve tests
ivanmorozov333 Dec 24, 2024
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
109 changes: 83 additions & 26 deletions ydb/core/kqp/ut/olap/indexes_ut.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -302,9 +302,8 @@ Y_UNIT_TEST_SUITE(KqpOlapIndexes) {
}

AFL_VERIFY(updatesCount + 6 ==
(ui64)csController->GetActualizationRefreshSchemeCount().Val())(
"updates", updatesCount)("count",
csController->GetActualizationRefreshSchemeCount().Val());
(ui64)csController->GetActualizationRefreshSchemeCount().Val())("updates", updatesCount)(
"count", csController->GetActualizationRefreshSchemeCount().Val());
}

class TTestIndexesScenario {
Expand Down Expand Up @@ -340,6 +339,17 @@ Y_UNIT_TEST_SUITE(KqpOlapIndexes) {
auto alterResult = session.ExecuteSchemeQuery(alterQuery).GetValueSync();
UNIT_ASSERT_VALUES_EQUAL_C(alterResult.GetStatus(), NYdb::EStatus::SUCCESS, alterResult.GetIssues().ToString());
}
{
auto alterQuery =
TStringBuilder() << Sprintf(
R"(ALTER OBJECT `/Root/olapStore` (TYPE TABLESTORE) SET (ACTION=UPSERT_INDEX, NAME=index_ngramm_uid, TYPE=BLOOM_NGRAMM_FILTER,
FEATURES=`{"column_name" : "resource_id", "ngramm_size" : 3, "hashes_count" : 2, "filter_size_bytes" : 64024}`);
)",
StorageId.data());
auto session = tableClient.CreateSession().GetValueSync().GetSession();
auto alterResult = session.ExecuteSchemeQuery(alterQuery).GetValueSync();
UNIT_ASSERT_VALUES_EQUAL_C(alterResult.GetStatus(), NYdb::EStatus::SUCCESS, alterResult.GetIssues().ToString());
}
{
auto alterQuery =
TStringBuilder() << Sprintf(
Expand Down Expand Up @@ -416,6 +426,27 @@ Y_UNIT_TEST_SUITE(KqpOlapIndexes) {
// important checker for control compactions (<=21) and control indexes constructed (>=21)
AFL_VERIFY(csController->GetCompactionStartedCounter().Val() == 21)("count", csController->GetCompactionStartedCounter().Val());

{
auto it = tableClient
.StreamExecuteScanQuery(R"(
--!syntax_v1

SELECT
COUNT(*)
FROM `/Root/olapStore/olapTable`
WHERE resource_id LIKE '%110a151%'
)")
.GetValueSync();

UNIT_ASSERT_C(it.IsSuccess(), it.GetIssues().ToString());
TString result = StreamResultToYson(it);
AFL_WARN(NKikimrServices::TX_COLUMNSHARD)("result", result);
AFL_WARN(NKikimrServices::TX_COLUMNSHARD)("skip", csController->GetIndexesSkippingOnSelect().Val())(
"check", csController->GetIndexesApprovedOnSelect().Val());
CompareYson(result, R"([[0u;]])");
AFL_VERIFY(!csController->GetIndexesApprovedOnSelect().Val());
AFL_VERIFY(csController->GetIndexesSkippingOnSelect().Val());
}
{
auto it = tableClient
.StreamExecuteScanQuery(R"(
Expand All @@ -441,30 +472,56 @@ Y_UNIT_TEST_SUITE(KqpOlapIndexes) {
}
AFL_VERIFY(csController->GetIndexesApprovedOnSelect().Val() < csController->GetIndexesSkippingOnSelect().Val());
}
ui32 requestsCount = 100;
for (ui32 i = 0; i < requestsCount; ++i) {
const ui32 idx = RandomNumber<ui32>(uids.size());
const auto query = [](const TString& res, const TString& uid, const ui32 level) {
TStringBuilder sb;
sb << "SELECT COUNT(*) FROM `/Root/olapStore/olapTable`" << Endl;
sb << "WHERE(" << Endl;
sb << "resource_id = '" << res << "' AND" << Endl;
sb << "uid= '" << uid << "' AND" << Endl;
sb << "level= " << level << Endl;
sb << ")";
return sb;
};
auto it = tableClient.StreamExecuteScanQuery(query(resourceIds[idx], uids[idx], levels[idx])).GetValueSync();

UNIT_ASSERT_C(it.IsSuccess(), it.GetIssues().ToString());
TString result = StreamResultToYson(it);
Cout << csController->GetIndexesSkippingOnSelect().Val() << " / " << csController->GetIndexesApprovedOnSelect().Val() << " / "
<< csController->GetIndexesSkippedNoData().Val() << Endl;
CompareYson(result, R"([[1u;]])");
{
ui32 requestsCount = 100;
for (ui32 i = 0; i < requestsCount; ++i) {
const ui32 idx = RandomNumber<ui32>(uids.size());
const auto query = [](const TString& res, const TString& uid, const ui32 level) {
TStringBuilder sb;
sb << "SELECT COUNT(*) FROM `/Root/olapStore/olapTable`" << Endl;
sb << "WHERE(" << Endl;
sb << "resource_id = '" << res << "' AND" << Endl;
sb << "uid= '" << uid << "' AND" << Endl;
sb << "level= " << level << Endl;
sb << ")";
return sb;
};
auto it = tableClient.StreamExecuteScanQuery(query(resourceIds[idx], uids[idx], levels[idx])).GetValueSync();

UNIT_ASSERT_C(it.IsSuccess(), it.GetIssues().ToString());
TString result = StreamResultToYson(it);
Cout << csController->GetIndexesSkippingOnSelect().Val() << " / " << csController->GetIndexesApprovedOnSelect().Val()
<< " / " << csController->GetIndexesSkippedNoData().Val() << Endl;
CompareYson(result, R"([[1u;]])");
}
AFL_VERIFY(csController->GetIndexesApprovedOnSelect().Val() * 5 < csController->GetIndexesSkippingOnSelect().Val())
("approved", csController->GetIndexesApprovedOnSelect().Val())("skipped", csController->GetIndexesSkippingOnSelect().Val());
}
{
const ui64 skipStart = csController->GetIndexesSkippingOnSelect().Val();
const ui64 approveStart = csController->GetIndexesApprovedOnSelect().Val();
ui32 requestsCount = 100;
for (ui32 i = 0; i < requestsCount; ++i) {
const ui32 idx = RandomNumber<ui32>(uids.size());
const auto query = [](const TString& res, const TString& uid, const ui32 level) {
TStringBuilder sb;
sb << "SELECT COUNT(*) FROM `/Root/olapStore/olapTable`" << Endl;
sb << "WHERE" << Endl;
sb << "resource_id LIKE '%" << res << "%'" << Endl;
return sb;
};
auto it = tableClient.StreamExecuteScanQuery(query(resourceIds[idx], uids[idx], levels[idx])).GetValueSync();

UNIT_ASSERT_C(it.IsSuccess(), it.GetIssues().ToString());
TString result = StreamResultToYson(it);
Cout << resourceIds[idx] << "/" << csController->GetIndexesSkippingOnSelect().Val() - skipStart << " / "
<< csController->GetIndexesApprovedOnSelect().Val() - approveStart << " / "
<< csController->GetIndexesSkippedNoData().Val() << Endl;
CompareYson(result, R"([[1u;]])");
}
AFL_VERIFY(csController->GetIndexesSkippingOnSelect().Val() - skipStart > 1)("approved", csController->GetIndexesApprovedOnSelect().Val() - approveStart)(
"skipped", csController->GetIndexesSkippingOnSelect().Val() - skipStart);
}

AFL_VERIFY(csController->GetIndexesApprovedOnSelect().Val() * 5 < csController->GetIndexesSkippingOnSelect().Val())
("approved", csController->GetIndexesApprovedOnSelect().Val())("skipped", csController->GetIndexesSkippingOnSelect().Val());
}
};

Expand Down
16 changes: 16 additions & 0 deletions ydb/core/protos/flat_scheme_op.proto
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -391,6 +391,13 @@ message TRequestedBloomFilter {
repeated string ColumnNames = 3;
}

message TRequestedBloomNGrammFilter {
optional uint32 NGrammSize = 1;
optional uint32 FilterSizeBytes = 2;
optional uint32 HashesCount = 3;
optional string ColumnName = 4;
}

message TRequestedMaxIndex {
optional string ColumnName = 1;
}
Expand All @@ -410,6 +417,7 @@ message TOlapIndexRequested {
TRequestedBloomFilter BloomFilter = 40;
TRequestedMaxIndex MaxIndex = 41;
TRequestedCountMinSketch CountMinSketch = 42;
TRequestedBloomNGrammFilter BloomNGrammFilter = 43;
}
}

Expand All @@ -419,6 +427,13 @@ message TBloomFilter {
repeated uint32 ColumnIds = 3;
}

message TBloomNGrammFilter {
optional uint32 NGrammSize = 1;
optional uint32 FilterSizeBytes = 2;
optional uint32 HashesCount = 3;
optional uint32 ColumnId = 4;
}

message TMaxIndex {
optional uint32 ColumnId = 1;
}
Expand All @@ -441,6 +456,7 @@ message TOlapIndexDescription {
TBloomFilter BloomFilter = 41;
TMaxIndex MaxIndex = 42;
TCountMinSketch CountMinSketch = 43;
TBloomNGrammFilter BloomNGrammFilter = 44;
}
}

Expand Down
209 changes: 0 additions & 209 deletions ydb/core/protos/ssa.proto
View file
Open in desktop

This file was deleted.

Loading
Loading