This is an automated email from the ASF dual-hosted git repository.
dataroaring pushed a commit to branch branch-3.0
in repository https://gitbox.apache.org/repos/asf/doris.git
The following commit(s) were added to refs/heads/branch-3.0 by this push:
new efe5b147814 [opt](index compaction)Avoid get file size when create
index reader and remove unnecessary file exists (#41079)
efe5b147814 is described below
commit efe5b147814a2612641c67f1f2fee6e17d97e893
Author: qiye <[email protected]>
AuthorDate: Sun Sep 22 17:10:48 2024 +0800
[opt](index compaction)Avoid get file size when create index reader and
remove unnecessary file exists (#41079)
Get file size and file exists operations are very expensive in object
storage.
Index compaction may have plenty of small files, and the head operation
will be a lot time consumption.
---
be/src/olap/compaction.cpp | 90 +++++++++++++++++++++-------------------------
1 file changed, 41 insertions(+), 49 deletions(-)
diff --git a/be/src/olap/compaction.cpp b/be/src/olap/compaction.cpp
index d9ca73b88a1..a1278c58d19 100644
--- a/be/src/olap/compaction.cpp
+++ b/be/src/olap/compaction.cpp
@@ -627,7 +627,8 @@ Status Compaction::do_inverted_index_compaction() {
auto seg_path = DORIS_TRY(rowset->segment_path(seg_id));
auto inverted_index_file_reader =
std::make_unique<InvertedIndexFileReader>(
fs, std::string
{InvertedIndexDescriptor::get_index_file_path_prefix(seg_path)},
- _cur_tablet_schema->get_inverted_index_storage_format());
+ _cur_tablet_schema->get_inverted_index_storage_format(),
+ rowset->rowset_meta()->inverted_index_file_info(seg_id));
bool open_idx_file_cache = false;
RETURN_NOT_OK_STATUS_WITH_WARN(
inverted_index_file_reader->init(config::inverted_index_read_buffer_size,
@@ -840,56 +841,47 @@ void
Compaction::construct_skip_inverted_index(RowsetWriterContext& ctx) {
return false;
}
- auto inverted_index_file_reader =
std::make_unique<InvertedIndexFileReader>(
- fs,
- std::string {
-
InvertedIndexDescriptor::get_index_file_path_prefix(*seg_path)},
-
_cur_tablet_schema->get_inverted_index_storage_format());
- bool open_idx_file_cache = false;
- auto st =
inverted_index_file_reader->init(config::inverted_index_read_buffer_size,
-
open_idx_file_cache);
- if (!st.ok()) {
- LOG(WARNING) << "init index "
- <<
inverted_index_file_reader->get_index_file_path(index_meta)
- << " error:" << st;
- return false;
- }
-
- bool exists = false;
- if (!inverted_index_file_reader->index_file_exist(index_meta,
&exists).ok()) {
- LOG(ERROR) <<
inverted_index_file_reader->get_index_file_path(index_meta)
- << " fs->exists error";
- return false;
- }
-
- if (!exists) {
- LOG(WARNING) << "tablet[" << _tablet->tablet_id() << "]
column_unique_id["
- << col_unique_id << "],"
- <<
inverted_index_file_reader->get_index_file_path(index_meta)
- << " is not exists, will skip index
compaction";
- return false;
- }
+ std::string index_file_path;
+ try {
+ auto inverted_index_file_reader =
std::make_unique<InvertedIndexFileReader>(
+ fs,
+ std::string {
+
InvertedIndexDescriptor::get_index_file_path_prefix(*seg_path)},
+
_cur_tablet_schema->get_inverted_index_storage_format(),
+
rowset->rowset_meta()->inverted_index_file_info(i));
+ bool open_idx_file_cache = false;
+ auto st = inverted_index_file_reader->init(
+ config::inverted_index_read_buffer_size,
open_idx_file_cache);
+ index_file_path =
inverted_index_file_reader->get_index_file_path(index_meta);
+ if (!st.ok()) {
+ LOG(WARNING) << "init index " << index_file_path << "
error:" << st;
+ return false;
+ }
- // check index meta
- auto result = inverted_index_file_reader->open(index_meta);
- if (!result.has_value()) {
- LOG(WARNING) << "open index "
- <<
inverted_index_file_reader->get_index_file_path(index_meta)
- << " error:" << result.error();
- return false;
- }
- auto reader = std::move(result.value());
- std::vector<std::string> files;
- reader->list(&files);
- reader->close();
-
- // why is 3?
- // bkd index will write at least 3 files
- if (files.size() < 3) {
+ // check index meta
+ auto result = inverted_index_file_reader->open(index_meta);
+ if (!result.has_value()) {
+ LOG(WARNING)
+ << "open index " << index_file_path << "
error:" << result.error();
+ return false;
+ }
+ auto reader = std::move(result.value());
+ std::vector<std::string> files;
+ reader->list(&files);
+ reader->close();
+
+ // why is 3?
+ // bkd index will write at least 3 files
+ if (files.size() < 3) {
+ LOG(WARNING) << "tablet[" << _tablet->tablet_id() <<
"] column_unique_id["
+ << col_unique_id << "]," <<
index_file_path
+ << " is corrupted, will skip index
compaction";
+ return false;
+ }
+ } catch (CLuceneError& err) {
LOG(WARNING) << "tablet[" << _tablet->tablet_id() << "]
column_unique_id["
- << col_unique_id << "],"
- <<
inverted_index_file_reader->get_index_file_path(index_meta)
- << " is corrupted, will skip index
compaction";
+ << col_unique_id << "] open index[" <<
index_file_path
+ << "], will skip index compaction, error:" <<
err.what();
return false;
}
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
