OrchANN: A Unified I/O Orchestration Framework for Skewed Out-of-Core Vector Search

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Huan, Chengying, Chen, Lizheng, Yang, Zhengyi, Ma, Shaonan, Gu, Rong, Yao, Renjie, Wang, Zhibin, Zhang, Mingxing, Xi, Fang, Tao, Jie, Zhang, Gang, Chen, Guihai, Tian, Chen
Format:	Preprint
Published:	2025
Subjects:	Databases Information Retrieval
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866915696848404480
author	Huan, Chengying Chen, Lizheng Yang, Zhengyi Ma, Shaonan Gu, Rong Yao, Renjie Wang, Zhibin Zhang, Mingxing Xi, Fang Tao, Jie Zhang, Gang Chen, Guihai Tian, Chen
author_facet	Huan, Chengying Chen, Lizheng Yang, Zhengyi Ma, Shaonan Gu, Rong Yao, Renjie Wang, Zhibin Zhang, Mingxing Xi, Fang Tao, Jie Zhang, Gang Chen, Guihai Tian, Chen
contents	Approximate nearest neighbor search (ANNS) at billion scale is fundamentally an out-of-core problem: vectors and indexes live on SSD, so performance is dominated by I/O rather than compute. Under skewed semantic embeddings, existing out-of-core systems break down: a uniform local index mismatches cluster scales, static routing misguides queries and inflates the number of probed partitions, and pruning is incomplete at the cluster level and lossy at the vector level, triggering "fetch-to-discard" reranking on raw vectors. We present OrchANN, an out-of-core ANNS engine that uses an I/O orchestration model for unified I/O governance along the route-access-verify pipeline. OrchANN selects a heterogeneous local index per cluster via offline auto-profiling, maintains a query-aware in-memory navigation graph that adapts to skewed workloads, and applies multi-level pruning with geometric bounds to filter both clusters and vectors before issuing SSD reads. Across five standard datasets under strict out-of-core constraints, OrchANN outperforms four baselines including DiskANN, Starling, SPANN, and PipeANN in both QPS and latency while reducing SSD accesses. Furthermore, OrchANN delivers up to 17.2x higher QPS and 25.0x lower latency than competing systems without sacrificing accuracy.
format	Preprint
id	arxiv_https___arxiv_org_abs_2512_22838
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	OrchANN: A Unified I/O Orchestration Framework for Skewed Out-of-Core Vector Search Huan, Chengying Chen, Lizheng Yang, Zhengyi Ma, Shaonan Gu, Rong Yao, Renjie Wang, Zhibin Zhang, Mingxing Xi, Fang Tao, Jie Zhang, Gang Chen, Guihai Tian, Chen Databases Information Retrieval Approximate nearest neighbor search (ANNS) at billion scale is fundamentally an out-of-core problem: vectors and indexes live on SSD, so performance is dominated by I/O rather than compute. Under skewed semantic embeddings, existing out-of-core systems break down: a uniform local index mismatches cluster scales, static routing misguides queries and inflates the number of probed partitions, and pruning is incomplete at the cluster level and lossy at the vector level, triggering "fetch-to-discard" reranking on raw vectors. We present OrchANN, an out-of-core ANNS engine that uses an I/O orchestration model for unified I/O governance along the route-access-verify pipeline. OrchANN selects a heterogeneous local index per cluster via offline auto-profiling, maintains a query-aware in-memory navigation graph that adapts to skewed workloads, and applies multi-level pruning with geometric bounds to filter both clusters and vectors before issuing SSD reads. Across five standard datasets under strict out-of-core constraints, OrchANN outperforms four baselines including DiskANN, Starling, SPANN, and PipeANN in both QPS and latency while reducing SSD accesses. Furthermore, OrchANN delivers up to 17.2x higher QPS and 25.0x lower latency than competing systems without sacrificing accuracy.
title	OrchANN: A Unified I/O Orchestration Framework for Skewed Out-of-Core Vector Search
topic	Databases Information Retrieval
url	https://arxiv.org/abs/2512.22838