Skip to content

docs: 补齐本轮优化的迭代复盘(8 篇迭代记录 + 1 篇测量纪律) - #302

Merged
NeverENG merged 1 commit into
mainfrom
docs/iteration-records
Aug 13, 2026
Merged

docs: 补齐本轮优化的迭代复盘(8 篇迭代记录 + 1 篇测量纪律)#302
NeverENG merged 1 commit into
mainfrom
docs/iteration-records

Conversation

@NeverENG

Copy link
Copy Markdown
Owner

本轮所有优化的量化依据与修法理由此前只存在于 PR 描述里,按 docs/ 既有「一篇一事」格式落成文档。

迭代记录

文档 结论
read-path-syscalls 常驻句柄 + 单次整块读 + 块缓存 + 去分配:点查 12.6k → 18.7 万 QPS
write-path-attribution 一个拼错的配置键吃掉 4.83× 写吞吐;分片 WAL 被实测证伪(0.23×)
silent-data-loss 故障注入测出四处静默丢数据,含故障注入手法表
data-races 两处数据竞争,均在最热路径上
scan-covers-sstable 扫描曾只看内存表,下游投递漏投全部已落盘记录
retention 按已投递位点丢弃整份文件,及其六条保守约束的理由
iterator-buffered-read 一层缓冲读换来顺序扫描 24×、八路归并 17.4×
structure-and-sdk 包边界、依赖注入、对外 SDK、压测脚本此前跑不起来

测量纪律

docs/measurement-discipline.md 收录 9 条被真实误判换来的规矩:交替 A/B 抵消热漂移、压测口径要确认压到目标路径、先归因再优化、「优化无效」与「优化没生效」不可混淆、变异测试、故障注入的可移植手法、-race 是并发缺陷的必要条件、E2E 先怀疑自己、grep 计数要复核。

每篇都保留了「后续」里未解决的问题(投递游标按 key 序的语义边界、Raft WAL 缺故障注入、service 层未参数化等),不粉饰。

纯文档改动,无代码变更。

🤖 Generated with Claude Code

按 docs/ 既有「一篇一事」格式补 8 篇迭代记录与 1 篇测量纪律,把本轮每项优化的
量化依据、修法理由与遗留问题落到文档,避免只存在于 PR 描述里。

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
@coderabbitai

coderabbitai Bot commented Aug 13, 2026

Copy link
Copy Markdown

Warning

Review limit reached

@NeverENG, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 33 minutes

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details
⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 375d8e81-ddef-4107-b9f3-95bd4944816c

📥 Commits

Reviewing files that changed from the base of the PR and between 1561b14 and 7e27d12.

📒 Files selected for processing (9)
  • docs/iteration-2026-08-13-data-races.md
  • docs/iteration-2026-08-13-iterator-buffered-read.md
  • docs/iteration-2026-08-13-read-path-syscalls.md
  • docs/iteration-2026-08-13-retention.md
  • docs/iteration-2026-08-13-scan-covers-sstable.md
  • docs/iteration-2026-08-13-silent-data-loss.md
  • docs/iteration-2026-08-13-structure-and-sdk.md
  • docs/iteration-2026-08-13-write-path-attribution.md
  • docs/measurement-discipline.md

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

@NeverENG
NeverENG merged commit 5d9759d into main Aug 13, 2026
3 of 4 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant