opensearch-project/OpenSearch

[Feature Request] Enable debug logging for Approximation framework

Open

#18,318 创建于 2025年5月16日

在 GitHub 查看
 (7 评论) (0 反应) (2 负责人)Java (8,123 star) (1,505 fork)batch import
Search:Performanceenhancementgood first issue

描述

Is your feature request related to a problem? Please describe

Similar to how we have debug logs in other optimization paths eg - https://github.com/opensearch-project/OpenSearch/blob/5fb4e6951b6c80db5c9c45398ed3704ac4092ba3/server/src/main/java/org/opensearch/search/aggregations/bucket/filterrewrite/FilterRewriteOptimizationContext.java#L89.

We should enable debug logging for how the optimization works, how many docs per segment we are collecting. How many nodes of the point tree we are visiting, how many leaves we are reading etc. This should help us understand if/when there are regressions.

Describe the solution you'd like

  • Enable debug logging
  • Enable more logs by adding them to AbstractProfiler so that we can profile the results

Related component

Search:Performance

Describe alternatives you've considered

No response

Additional context

No response

贡献者指南

技术栈
java
领域
performance
议题类型
feature
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
3
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
half day
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
blocked
清晰度议题是否清楚说明期望改动、验收标准和下一步。
clear
前置要求
Java knowledgeOpenSearch codebase familiarity
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
35
研究方向
The issue requests debug logging for the Approximation framework. Start by examining the existing debug logging in FilterRewriteOptimizationContext.java (linked in the issue) to understand the pattern. Then locate the Approximation framework's optimization classes and AbstractProfiler to add similar logging. Consider measuring docs per segment, point tree nodes visited, and leaves read. Since there are two assignees, check the issue comments for any discussion on implementation details.