arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2601.08528cs.DB
收藏

SVFusion:一种用于大规模实时向量搜索的CPU-GPU协同架构

SVFusion: A CPU-GPU Co-Processing Architecture for Large-Scale Real-Time Vector Search

Zhejiang University(浙江大学);Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security(杭州高新技术产业开发区(滨江)区块链与数据安全研究所)

Yuchen Peng, Dingyu Yang, Zhongle Xie, Ji Sun, Lidan Shou, Ke Chen, Gang Chen

AI总结:

SVFusion通过CPU-GPU协同处理和动态更新机制,提升大规模实时向量搜索的性能和效率。

摘要:

近似最近邻搜索(ANNS)是信息检索和推荐等现代应用的基础。随着向量数据的快速增长,高效索引以实现实时向量搜索已成为基本需求。现有基于CPU的解决方案支持更新,但吞吐量低;而基于GPU的系统具有高性能,但面临动态更新和有限GPU内存的挑战,导致需要同时准确和快速的连续大规模向量搜索存在性能差距。本文提出SVFusion,一种用于实时向量搜索的GPU-CPU-磁盘协同框架,将复杂的GPU计算与在线更新相结合。SVFusion利用分层向量索引架构,采用CPU-GPU协同处理,以及工作负载感知的向量缓存机制,以最大化有限GPU内存的效率。它进一步通过实时协调CUDA多流优化和自适应资源管理,以及并发控制,确保在交错查询和更新下数据的一致性。实验结果表明,SVFusion在查询延迟和吞吐量上实现了显著改进,平均吞吐量比基线方法高20.9倍,延迟降低1.3倍至50.7倍,同时在各种流工作负载下保持高召回率。

英文摘要:

Approximate Nearest Neighbor Search (ANNS) underpins modern applications such as information retrieval and recommendation. With the rapid growth of vector data, efficient indexing for real-time vector search has become rudimentary. Existing CPU-based solutions support updates but suffer from low throughput, while GPU-accelerated systems deliver high performance but face challenges with dynamic updates and limited GPU memory, resulting in a critical performance gap for continuous, large-scale vector search requiring both accuracy and speed. In this paper, we present SVFusion, a GPU-CPU-disk collaborative framework for real-time vector search that bridges sophisticated GPU computation with online updates. SVFusion leverages a hierarchical vector index architecture that employs CPU-GPU co-processing, along with a workload-aware vector caching mechanism to maximize the efficiency of limited GPU memory. It further enhances performance through real-time coordination with CUDA multi-stream optimization and adaptive resource management, along with concurrency control that ensures data consistency under interleaved queries and updates. Empirical results demonstrate that SVFusion achieves significant improvements in query latency and throughput, exhibiting a 20.9x higher throughput on average and 1.3x to 50.7x lower latency compared to baseline methods, while maintaining high recall for large-scale datasets under various streaming workloads.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑