YoVinchen
ec1acb36b5
Some checks failed
secret-scan / scan (push) Has been cancelled
The project begins with architecture, data governance, reproducible evaluation, deployment boundaries, and secret-handling contracts so later code has measurable acceptance criteria. Constraint: Real provider credentials, workspace identities, and restricted geological data must never enter Git Rejected: Add placeholder runnable services in the design commit | would imply unverified implementation readiness Confidence: high Scope-risk: narrow Reversibility: clean Directive: Run make verify before every commit and update ADRs when architecture boundaries change Tested: Secret scan, Markdown links, YAML parse, shell syntax, and staged diff validation Not-tested: Application runtime is intentionally deferred to the implementation stage
26 lines
1.4 KiB
Markdown
26 lines
1.4 KiB
Markdown
# ADR-0001：第一版使用 PostgreSQL + pgvector
|
||
|
||
- Status: accepted
|
||
- Date: 2026-07-11
|
||
|
||
## Context
|
||
|
||
系统目标是毕业设计规模的地质文档 RAG，并要求 Docker 最简单启动、数据和实验可复现。预计正式语料约 1 万至 30 万切片，默认向量维度为 1024。
|
||
|
||
## Decision
|
||
|
||
第一版使用 PostgreSQL + pgvector，同时存储业务元数据和向量。使用 Cosine 距离并从首个 schema 版本建立 HNSW；评测事务通过禁用索引扫描获得同过滤条件的 exact 基线。应用与数据库之外不增加 Qdrant、Milvus、Redis 等基础服务，后台任务使用 PostgreSQL 持久化任务表。
|
||
|
||
## Consequences
|
||
|
||
- Docker 长期运行组件减少，备份与事务边界更清晰。
|
||
- 1024 维位于 pgvector HNSW `vector` 类型 2000 维限制内。
|
||
- 带权限过滤的 ANN 必须使用冗余过滤列、普通索引和 pgvector 0.8.x iterative scan，并与 exact filtered Top K 比较召回。
|
||
- 维度变化需要新表、列或索引，不允许把不同维度混写。
|
||
- 当切片达到百万级、过滤复杂度或查询吞吐明显超过单机 PostgreSQL 能力时，重新评估 Qdrant；Milvus 只在明确需要分布式大规模检索时评估。
|
||
|
||
## Rejected
|
||
|
||
- Qdrant：RAG 能力很好，但第一版会引入第二套持久化系统和双写一致性问题。
|
||
- Milvus：组件和运维成本超过当前规模需求。
|