Technology Radar
Published : Apr 02, 2025
NOT ON THE CURRENT EDITION
This blip is not on the current edition of the Radar. If it was on one of the last few editions, it is likely that it is still relevant. If the blip is older, it might no longer be relevant and our assessment might be different today. Unfortunately, we simply don't have the bandwidth to continuously review blips from previous editions of the Radar.
Understand more
Apr 2025
Assess
turbopuffer 是一个无服务器的多租户搜索引擎,能够在对象存储上无缝集成向量搜索和全文搜索。我们非常欣赏其架构和设计选择,尤其是在耐久性、可扩展性和成本效率方面的专注。通过将对象存储用作预写日志并保持查询节点的无状态化,它非常适合高规模的搜索工作负载。
turbopuffer 专为性能和准确性而设计,开箱即可提供高召回率,即使是复杂的基于过滤条件的查询也不例外。它将冷查询结果缓存到 NVMe SSD,并将经常访问的命名空间保存在内存中,从而实现对数十亿文档的低延迟搜索。这使其非常适合大规模文档检索、向量搜索以及检索增强生成(RAG)等 AI 应用。然而,其对对象存储的依赖也带来了查询延迟的权衡,使其在需要无状态分布式计算的工作负载中最为高效。turbopuffer 驱动了诸如 Cursor 这样的大规模生产系统,但目前仅通过推荐或邀请方式获得访问权限。