构建向量嵌入存储

项目概述

本项目实现一个面向 vector similarity search 的完整系统，核心要解决的问题是 finding the k nearest neighbors over millions of floating-point vectors without scanning every one。整体由 HNSW index、Distance metric、Insert pipeline、Search reranking、Metadata filter 五个模块组成，每个模块单一职责、可独立测试，最后通过集成测试验证端到端正确性。设计上把 a high insert rate that degrades the graph before a background compaction can re-link nodes 作为第一类需求先行处理，而非事后再补。

架构与关键决策

数据模型为 a flat array of fixed-dimension float vectors plus a graph overlay of neighbor links per node。HNSW index 定义其余阶段共同依赖的数据形状并在边界处校验输入；Distance metric 承载 vector similarity search 的正确性逻辑；Insert pipeline 专门处理 a high insert rate that degrades the graph before a background compaction can re-link nodes，在失败路径上给出明确的拒绝、重试或降级策略；Search reranking 提供可观测性，每次操作输出带关联 id 的结构化事件；Metadata filter 在过载时按既定方式降级，避免过载演变为数据损坏。配置与代码分离，使风险变更可经 diff 审阅。

实施与测试

按五个阶段依次实现 HNSW index、Distance metric、Insert pipeline、Search reranking、Metadata filter，每阶段配套聚焦单测。集成测试组合各模块并驱动真实 vector similarity search 场景：先验证正常路径，再刻意触发 a high insert rate that degrades the graph before a background compaction can re-link nodes 并断言系统降级而非损坏。所有依赖均在进程内或被桩替换，用例在 CI 中确定可复现。

限制与后续

• 单区域设计，多区域复制超出范围。
• HNSW index 默认进程内，未水平共享。
• 无鉴权层，假定部署在强制身份的边缘之后。

代码示例

# setup for vector-embedding-store
pnpm init
pnpm add hnswlib-node fastify pino
pnpm add -D typescript tsx vitest

vector_embedding_store/
├── src/index.ts
├── src/hnsw-index.ts
├── src/distance-metric.ts
├── tests/vector_embedding_store.test.ts
└── package.json

// HNSW index: vector similarity search core
export function hnswindex(input: HnswindexInput): HnswindexResult {
  // Build the hnsw index that anchors vector similarity search and defines the core data shape.
  return processHnswindex(input);
}

// Distance metric — handles vector similarity search Add the distance metric so
export class Distancemetric {
  constructor(private opts: DistancemetricOptions) {}
  async run(ctx: Context): Promise<DistancemetricOutput> {
    return this.execute(ctx);
  }
}

// Insert pipeline: vector similarity search stage
export const insertpipeline = (next: Next) => async (req: Request) => {
  const out = await transform(req);
  return next(out);
};

// Search reranking for vector similarity search
export async function searchreranking(stream: AsyncIterable<Chunk>): Promise<Aggregate> {
  let acc = init();
  for await (const c of stream) acc = merge(acc, c);
  return acc;
}

// Metadata filter: vector similarity search finalization
export function metadatafilter(state: State): Result {
  validate(state);
  return finalize(state);
}

# run the vector similarity search test suite
pnpm vitest run