[Enhancement] Constant LIKE/REGEXP pattern is recompiled for every clone of the expression (per scanner and per segment iterator)

未关闭
  • 哪里不对？
#69,916 1 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 1 天内回复

还没有人认领这个 Issue。

评估

难度
4/5
预计耗时
3-5 天
新手友好度
45/100
Issue 类型
功能
描述清晰度
基本清楚
活跃度
活跃
技术栈
cpp, sql
领域
databases, performance

调研方向

Start from FunctionLike and FunctionRegexpLike in the backend expr code, where THREAD_LOCAL open compiles the pattern. Read VExprContext::clone and SegmentIterator::_construct_compound_expr_context to see per-clone opens. Implement FRAGMENT_LOCAL compilation storing state in FunctionContext, then reuse via hs_clone_scratch. Validate with the provided REGEXP SQL on a many-segment table; done when open phase no longer scales with segments.

由索引模型根据 Issue 内容生成。

描述

Search before asking
  • I had searched in the issues and found no similar issues.
Version

master (checked at 5b2c066b4d), branch-4.1, 4.1.5

What's wrong?

FunctionLike and FunctionRegexpLike build their state, including the Hyperscan database (hs_compile) or the RE2/Boost regex for a constant pattern, only in their THREAD_LOCAL open. VExprContext::clone opens THREAD_LOCAL again for every clone, and a pushed-down predicate is cloned once per scanner and once per segment iterator (SegmentIterator::_construct_compound_expr_context). A scan over N segments therefore compiles the same constant pattern about N times.

A case-insensitive alternation such as (?i)(alpha|bravo|...|juliet)[.](exe|dll)$ takes milliseconds to compile. On a table with about ten thousand segments, the open phase (SegmentIteratorInitTimer in the profile) dominates the query and costs minutes of CPU. Matching itself (ExprFilterEvalTime) stays in the tens of milliseconds.

LikeSearchState::clone, used by LikeColumnPredicate, also runs hs_prepare again for each clone.

What you expected to happen

A constant pattern is compiled once per fragment, and its clones reuse it.

How to reproduce
CREATE TABLE t (k BIGINT, s STRING) DUPLICATE KEY(k) DISTRIBUTED BY HASH(k) BUCKETS 64
PROPERTIES ('replication_num' = '1', 'disable_auto_compaction' = 'true');
-- 60 times:
INSERT INTO t SELECT number, concat('file', cast(number AS STRING), '.exe') FROM numbers('number' = '5000');

SELECT count(*) FROM t WHERE s REGEXP 'file1[.]exe$';
SELECT count(*) FROM t WHERE s REGEXP '(?i)(alpha|bravo|charlie|delta|echo|foxtrot|golf|hotel|india|juliet|file1)[.](exe|dll)$';

With about 3.9k rowsets on one BE, the second query takes about 150 ms against about 72 ms for the first: about 77 ms of compiling. With the change below, both take about 47 ms. The gap grows linearly with the segment count.

Anything else?

Proposed change:

  • In the FRAGMENT_LOCAL open, compile a constant pattern and store it as fragment state. FunctionContext::clone() already shares that state.
  • In the THREAD_LOCAL open, share the fragment's Hyperscan database, which is read-only once compiled, and give each state its own scratch through hs_clone_scratch.
  • Keep compiling RE2 and Boost per state, because RE2 locks its lazily built DFA cache on every search, so one shared RE2 would serialise matching across scanner threads.
  • Non-constant patterns keep the current per-call path.
  • LikeSearchState::clone shares the database the same way.

Scope: this applies to LIKE/REGEXP, the only functions that compile a single pattern with Hyperscan. regexp_count, regexp_replace, regexp_extract and split_by_regexp also compile per THREAD_LOCAL open, but they use RE2. Building an RE2 is much cheaper than hs_compile, and sharing one RE2 across clones would contend on its DFA cache lock, so they are better left compiling per clone.

Related: #68977

Are you willing to submit PR?
  • Yes I am willing to submit a PR!
Code of Conduct
主要语言
Java
星标
16.2k
派生
4k
平均合并
1 天 16 小时
30 天内合并 PR
679

环境准备

从这里开始

  1. 先读完整个 Issue，再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库，在一个分支上完成修改。
  4. 提交 Pull Request，并在描述里引用这个 Issue 编号。

apache/doris 的其他 Issue

查看 apache/doris 的全部 Issue

相似的 Issue

更多 Java Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。