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.

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

[opt](inverted index) mow supports index optimization #37428

Merged
merged 1 commit into from
Jul 16, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
52 changes: 45 additions & 7 deletions be/src/olap/rowset/segment_v2/segment_iterator.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -2026,11 +2026,12 @@ Status SegmentIterator::_read_columns_by_index(uint32_t nrows_read_limit, uint32
auto debug_col_name = DebugPoints::instance()->get_debug_param_or_default<std::string>(
"segment_iterator._read_columns_by_index", "column_name", "");
if (debug_col_name.empty()) {
return Status::Error<ErrorCode::INTERNAL_ERROR>("{} does not need to read data");
return Status::Error<ErrorCode::INTERNAL_ERROR>("does not need to read data");
}
auto col_name = _opts.tablet_schema->column(cid).name();
if (debug_col_name.find(col_name) != std::string::npos) {
return Status::Error<ErrorCode::INTERNAL_ERROR>("{} does not need to read data");
return Status::Error<ErrorCode::INTERNAL_ERROR>("does not need to read data, {}",
debug_col_name);
}
})

Expand Down Expand Up @@ -2225,9 +2226,27 @@ Status SegmentIterator::_read_columns_by_rowids(std::vector<ColumnId>& read_colu
}

for (auto cid : read_column_ids) {
if (_prune_column(cid, (*mutable_columns)[cid], true, select_size)) {
auto& colunm = (*mutable_columns)[cid];
if (_no_need_read_key_data(cid, colunm, select_size)) {
Copy link
Contributor

Choose a reason for hiding this comment

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

can you call _no_need_read_key_data inside _prune_column?

continue;
}
if (_prune_column(cid, colunm, true, select_size)) {
continue;
}

DBUG_EXECUTE_IF("segment_iterator._read_columns_by_index", {
auto debug_col_name = DebugPoints::instance()->get_debug_param_or_default<std::string>(
"segment_iterator._read_columns_by_index", "column_name", "");
if (debug_col_name.empty()) {
return Status::Error<ErrorCode::INTERNAL_ERROR>("does not need to read data");
}
auto col_name = _opts.tablet_schema->column(cid).name();
if (debug_col_name.find(col_name) != std::string::npos) {
return Status::Error<ErrorCode::INTERNAL_ERROR>("does not need to read data, {}",
debug_col_name);
}
})

RETURN_IF_ERROR(_column_iterators[cid]->read_by_rowids(rowids.data(), select_size,
_current_return_columns[cid]));
}
Expand Down Expand Up @@ -2385,6 +2404,13 @@ Status SegmentIterator::_next_batch_internal(vectorized::Block* block) {
nrows_read_limit = std::min(static_cast<uint32_t>(_opts.topn_limit), nrows_read_limit);
}

DBUG_EXECUTE_IF("segment_iterator.topn_opt", {
if (nrows_read_limit != 1) {
return Status::Error<ErrorCode::INTERNAL_ERROR>("topn opt execute failed: {}",
nrows_read_limit);
}
})

RETURN_IF_ERROR(_init_current_block(block, _current_return_columns, nrows_read_limit));
_converted_column_ids.assign(_schema->columns().size(), 0);

Expand Down Expand Up @@ -2837,7 +2863,10 @@ bool SegmentIterator::_no_need_read_key_data(ColumnId cid, vectorized::MutableCo
if (_opts.runtime_state && !_opts.runtime_state->query_options().enable_no_need_read_data_opt) {
return false;
}
if (_opts.tablet_schema->keys_type() != KeysType::DUP_KEYS) {

if (!((_opts.tablet_schema->keys_type() == KeysType::DUP_KEYS ||
(_opts.tablet_schema->keys_type() == KeysType::UNIQUE_KEYS &&
_opts.enable_unique_key_merge_on_write)))) {
return false;
}

Expand Down Expand Up @@ -2895,11 +2924,20 @@ bool SegmentIterator::_can_opt_topn_reads() const {
return false;
}

if (!_col_predicates.empty() || !_col_preds_except_leafnode_of_andnode.empty()) {
return false;
std::set<uint32_t> cids;
for (auto* pred : _col_predicates) {
cids.insert(pred->column_id());
}
for (auto* pred : _col_preds_except_leafnode_of_andnode) {
cids.insert(pred->column_id());
}

return true;
uint32_t delete_sign_idx = _opts.tablet_schema->delete_sign_idx();
bool result = std::ranges::all_of(cids.begin(), cids.end(), [delete_sign_idx](auto cid) {
return cid == delete_sign_idx;
});

return result;
}

} // namespace segment_v2
Expand Down
2 changes: 2 additions & 0 deletions be/src/vec/exprs/vexpr.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -604,6 +604,8 @@ bool VExpr::fast_execute(Block& block, const ColumnNumbers& arguments, size_t re
size_t input_rows_count, const std::string& function_name) {
std::string result_column_name = gen_predicate_result_sign(block, arguments, function_name);
if (!block.has(result_column_name)) {
DBUG_EXECUTE_IF("segment_iterator.fast_execute",
{ return Status::Error<ErrorCode::INTERNAL_ERROR>("fast_execute failed"); })
return false;
}

Expand Down
Original file line number Diff line number Diff line change
@@ -0,0 +1,31 @@
-- This file is automatically generated. You should know what you did if you want to edit this
-- !sql --
210

-- !sql --
2

-- !sql --
4

-- !sql --
29

-- !sql --
14

-- !sql --
120

-- !sql --
2

-- !sql --
4

-- !sql --
22

-- !sql --
11

Original file line number Diff line number Diff line change
@@ -0,0 +1,25 @@
-- This file is automatically generated. You should know what you did if you want to edit this
-- !sql --
893964617 40.135.0.0 GET /images/hm_bg.jpg HTTP/1.0 200 24736

-- !sql --
893964672 26.1.0.0 GET /images/hm_bg.jpg HTTP/1.0 200 24736

-- !sql --
893964672 26.1.0.0 GET /images/hm_bg.jpg HTTP/1.0 200 24736

-- !sql --
893964672 26.1.0.0 GET /images/hm_bg.jpg HTTP/1.0 200 24736

-- !sql --
893964617 40.135.0.0 GET /images/hm_bg.jpg HTTP/1.0 200 24736

-- !sql --
893964672 26.1.0.0 GET /images/hm_bg.jpg HTTP/1.0 200 24736

-- !sql --
893964672 26.1.0.0 GET /images/hm_bg.jpg HTTP/1.0 200 24736

-- !sql --
893964672 26.1.0.0 GET /images/hm_bg.jpg HTTP/1.0 200 24736

Original file line number Diff line number Diff line change
@@ -0,0 +1,122 @@
// Licensed to the Apache Software Foundation (ASF) under one
// or more contributor license agreements. See the NOTICE file
// distributed with this work for additional information
// regarding copyright ownership. The ASF licenses this file
// to you under the Apache License, Version 2.0 (the
// "License"); you may not use this file except in compliance
// with the License. You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing,
// software distributed under the License is distributed on an
// "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
// KIND, either express or implied. See the License for the
// specific language governing permissions and limitations
// under the License.

suite("test_all_index_hit_fault_injection", "nonConcurrent") {
// define a sql table
def indexTbName1 = "test_all_index_hit_fault_injection_1"
def indexTbName2 = "test_all_index_hit_fault_injection_2"

sql "DROP TABLE IF EXISTS ${indexTbName1}"
sql """
CREATE TABLE ${indexTbName1} (
`@timestamp` int(11) NULL COMMENT "",
`clientip` varchar(20) NULL COMMENT "",
`request` text NULL COMMENT "",
`status` int(11) NULL COMMENT "",
`size` int(11) NULL COMMENT "",
INDEX clientip_idx (`clientip`) USING INVERTED COMMENT '',
INDEX request_idx (`request`) USING INVERTED PROPERTIES("parser" = "english", "support_phrase" = "true") COMMENT ''
) ENGINE=OLAP
DUPLICATE KEY(`@timestamp`)
COMMENT "OLAP"
DISTRIBUTED BY RANDOM BUCKETS 1
PROPERTIES (
"replication_allocation" = "tag.location.default: 1",
"disable_auto_compaction" = "true"
);
"""

sql "DROP TABLE IF EXISTS ${indexTbName2}"
sql """
CREATE TABLE ${indexTbName2} (
`@timestamp` int(11) NULL COMMENT "",
`clientip` varchar(20) NULL COMMENT "",
`request` text NULL COMMENT "",
`status` int(11) NULL COMMENT "",
`size` int(11) NULL COMMENT "",
INDEX clientip_idx (`clientip`) USING INVERTED COMMENT '',
INDEX request_idx (`request`) USING INVERTED PROPERTIES("parser" = "english", "support_phrase" = "true") COMMENT ''
) ENGINE=OLAP
UNIQUE KEY(`@timestamp`)
COMMENT "OLAP"
DISTRIBUTED BY HASH(`@timestamp`) BUCKETS 1
PROPERTIES (
"replication_allocation" = "tag.location.default: 1",
"enable_unique_key_merge_on_write" = "true",
"disable_auto_compaction" = "true"
);
"""

def load_httplogs_data = {table_name, label, read_flag, format_flag, file_name, ignore_failure=false,
expected_succ_rows = -1, load_to_single_tablet = 'true' ->

// load the json data
streamLoad {
table "${table_name}"

// set http request header params
set 'label', label + "_" + UUID.randomUUID().toString()
set 'read_json_by_line', read_flag
set 'format', format_flag
file file_name // import json file
time 10000 // limit inflight 10s
if (expected_succ_rows >= 0) {
set 'max_filter_ratio', '1'
}

// if declared a check callback, the default check condition will ignore.
// So you must check all condition
check { result, exception, startTime, endTime ->
if (ignore_failure && expected_succ_rows < 0) { return }
if (exception != null) {
throw exception
}
log.info("Stream load result: ${result}".toString())
def json = parseJson(result)
}
}
}

try {
load_httplogs_data.call(indexTbName1, 'test_all_index_hit_fault_injection_1', 'true', 'json', 'documents-1000.json')
load_httplogs_data.call(indexTbName2, 'test_all_index_hit_fault_injection_2', 'true', 'json', 'documents-1000.json')

sql "sync"

try {
GetDebugPoint().enableDebugPointForAllBEs("segment_iterator._read_columns_by_index", [column_name: "clientip,request"])
GetDebugPoint().enableDebugPointForAllBEs("segment_iterator.fast_execute")

qt_sql """ select count() from ${indexTbName1} where (request match_phrase 'hm'); """
qt_sql """ select count() from ${indexTbName1} where (request match_phrase 'hm' and clientip = '126.1.0.0'); """
qt_sql """ select count() from ${indexTbName1} where (request match_phrase 'hm' and clientip = '126.1.0.0') or (request match_phrase 'bg' and clientip = '201.0.0.0'); """
qt_sql """ select count() from ${indexTbName1} where (request match_phrase 'hm' and clientip = '126.1.0.0' or clientip = '247.37.0.0') or (request match_phrase 'bg' and clientip = '201.0.0.0' or clientip = '232.0.0.0'); """
qt_sql """ select count() from ${indexTbName1} where (request match_phrase 'hm' and clientip in ('126.1.0.0', '247.37.0.0')) or (request match_phrase 'bg' and clientip in ('201.0.0.0', '232.0.0.0')); """

qt_sql """ select count() from ${indexTbName2} where (request match_phrase 'hm'); """
qt_sql """ select count() from ${indexTbName2} where (request match_phrase 'hm' and clientip = '126.1.0.0'); """
qt_sql """ select count() from ${indexTbName2} where (request match_phrase 'hm' and clientip = '126.1.0.0') or (request match_phrase 'bg' and clientip = '201.0.0.0'); """
qt_sql """ select count() from ${indexTbName2} where (request match_phrase 'hm' and clientip = '126.1.0.0' or clientip = '247.37.0.0') or (request match_phrase 'bg' and clientip = '201.0.0.0' or clientip = '232.0.0.0'); """
qt_sql """ select count() from ${indexTbName2} where (request match_phrase 'hm' and clientip in ('126.1.0.0', '247.37.0.0')) or (request match_phrase 'bg' and clientip in ('201.0.0.0', '232.0.0.0')); """

} finally {
GetDebugPoint().disableDebugPointForAllBEs("segment_iterator._read_columns_by_index")
GetDebugPoint().disableDebugPointForAllBEs("segment_iterator.fast_execute")
}
} finally {
}
}
Loading
Loading