docs: 边缘/具身智能方向定位与可行性分析

[NeverENG]

内容

新增 docs/edge-positioning-feasibility.md：对"边缘/具身智能"转向方向的一次冷水复盘。

• 现状盘点：代码核实哪些卖点是真货（BanNet/钩子/LSM/Raft），哪些是 PPT（双通道"抢占式优先级调度"、MVCC）。
• 战场判断：边缘赛道真正的对手是 ROS2/DDS、rosbag2/MCAP、Zenoh、COS multipart，不是 Kafka；"轻量单二进制"是入场券不是护城河。
• 护城河体检：Multi-Raft 互备（工具用错）、去中心化调度（产品越界，砍）、时空索引+边缘切片上传（唯一真差异化，但未实现）。
• 路线图：停止换皮循环，做三件事——讲清内核深度、做实唯一差异化（钩子+边缘查询）、一个有数字的 IMU 高频 benchmark demo。
• 含外部事实核验（MCAP/Zenoh 在 ROS2 的现状，2026-06）。

仅新增一份文档，无代码改动。

🤖 Generated with Claude Code

Summary by CodeRabbit

Documentation

• Added comprehensive edge intelligence positioning and feasibility review, evaluating current implementation against strategic goals and competitive offerings.
• Included competitive landscape analysis and identified key differentiators around programmable network-layer hooks and edge-side filtering.
• Outlined prioritized technical recommendations for enhanced time-series API capabilities, scheduling improvements, and benchmark demonstrations with measurable performance metrics.

[coderabbitai]

Caution

Review failed

The pull request is closed.

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 6d80b53a-aae1-4bd0-b955-56e6cc059013

📥 Commits

Reviewing files that changed from the base of the PR and between fc34885 and ff11d59.

📒 Files selected for processing (1)

• docs/edge-positioning-feasibility.md

---

📝 Walkthrough

Walkthrough

This PR adds a comprehensive documentation file reviewing the project's edge/embodied intelligence positioning, competitive landscape, current implementation status, and proposing concrete next development steps including PreHandle filters, range-scan APIs with predicates, and benchmark demonstration requirements.

Changes

Edge Positioning & Feasibility Review

Layer / File(s)

Summary

Edge positioning and feasibility write-up docs/edge-positioning-feasibility.md

Complete positioning review article evaluating project positioning against four technical pillars, analyzing competitive landscape (ROS 2/DDS, Zenoh, rosbag2+MCAP, COS multipart), identifying core differentiators in programmable network-layer hooks with edge-side range filtering, critiquing proposed Multi-Raft and decentralized scheduling approaches, and proposing concrete roadmap: PreHandle filter, scan API with predicates, optional priority scheduling, and benchmark demo with measurable metrics.

Estimated Code Review Effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

Poem

A bunny hops through feasibility's gate,
Positioning clear, the vision's first-rate!
Edge hooks and slice filters, the path now laid bare,
Benchmarks and roadmaps—let's build with great care! 🐰✨

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch docs/edge-positioning-feasibility

---

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

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[github-actions]

🐯 BanGD 数据库内核评审

整体风险：🟢 低

变更总结：该 PR 仅新增一份产品/战略定位文档 docs/edge-positioning-feasibility.md，无任何代码改动。文档从数据库引擎架构师视角，对项目"边缘/具身智能"转向方向做了系统性复盘：逐条核验代码中真实实现 vs PPT 承诺（如双通道抢占式调度为空、MVCC 不存在）、重新定义对手盘（ROS2/DDS/MCAP/Zenoh 而非 Kafka）、逐项体检三大护城河的真实性与可行性、给出诚实的路线图与简历话术建议。文档本身是"架构评审级别的文字评审"，不影响任何代码路径、存储格式、并发模型或运行时行为。

本评审不阻塞合入；架构级建议以 Issue 形式跟踪，普通问题在下方内联列出。

未发现需要从架构层面改进的问题。

---

_{本次评审消耗 token：共 24143 tokens（输入 23876，输出 267，缓存命中 0，缓存写入 0）｜维度 [concurrency, memory, lock, storage, schema]｜对抗式复核 3 票/条，过滤疑似误报 0 条}