BatANN实现高效分布式向量搜索，突破单机存储瓶颈。
Passing the Baton: High Throughput Distributed Disk-Based Vector Search with BatANN
- 跨服务器查询时传递完整状态，提升数据局部性
- 10台服务器上10亿向量0.95召回率下延迟<3毫秒
- 开源支持单全局图，适用于超大规模检索场景
向量搜索是现代信息检索系统的核心，涵盖基于检索的生成（RAG）和非结构化文本与图像的搜索引擎。随着数据集规模达数十亿向量，基于磁盘的向量搜索成为实用方案。然而，未来数据量将超出单台服务器容量，本地固态硬盘吞吐量也难满足需求。我们提出BatANN，一种分布式磁盘型近似最近邻（ANN）系统，保持单个全局图的对数级搜索效率的同时，在服务器数量上实现近线性吞吐量扩展。核心创新在于：当访问其他机器存储的邻域时，将完整查询状态发送至目标机器继续执行，从而提升局部性。在10亿向量数据集、0.95召回率下，使用10台服务器时，BatANN相较散列-收集基线提升3.5-5.59倍吞吐量，比DistributedANN提升1.44-2.09倍，同时平均延迟低于3毫秒。所有结果均基于标准TCP实现。据我们所知，BatANN是首个开源的基于单全局图的分布式磁盘型向量搜索系统。
原文摘要 · Abstract (English)
Vector search underpins modern information-retrieval systems, including retrieval-augmented generation (RAG) pipelines and search engines over unstructured text and images. As datasets scale to billions of vectors, disk-based vector search has emerged as a practical solution. However, looking to the future, we must anticipate datasets too large for any single server and throughput demands that exceed the limits of locally attached SSDs. We present BatANN, a distributed disk-based approximate nearest neighbor (ANN) system that retains the logarithmic search efficiency of a single global graph while achieving near-linear throughput scaling in the number of servers. Our core innovation is that when accessing a neighborhood which is stored on another machine, we send the full state of the query to the other machine to continue executing there for improved locality. On 1B-point datasets at 0.95 recall using 10 servers, BatANN achieves 3.5-5.59x of the scatter-gather baseline and 1.44-2.09x the throughput of DistributedANN, respectively, while maintaining mean latency below 3 ms. Moreover, we get these results on standard TCP. To our knowledge, BatANN is the first open-source distributed disk-based vector search system to operate over a single global graph.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。