为华为鲲鹏CPU优化的高效向量搜索库，性能超主流x86方案。
KBest: Efficient Vector Search on Kunpeng CPU
- 针对鲲鹏920芯片设计，融合SIMD加速与预取等硬件级优化。
- 查询吞吐提升超2倍，实测优于运行在x86上的现有最优库。
- 已服务千万级日查询量，适合需低延迟向量检索的场景。
向量搜索从大规模向量数据集中返回与查询向量最相似的结果，支撑搜索、推荐及大模型等重要应用。为实现经济性，必须高效以降低资源消耗。然而，现有向量搜索库（如Faiss和DiskANN）主要针对x86 CPU架构优化，而华为鲲鹏CPU基于ARM架构，计算能力强劲。本文提出KBest，一个专为最新鲲鹏920 CPU设计的向量搜索库。通过大量硬件感知与算法优化，包括SIMD加速距离计算、数据预取、索引精炼、早期终止和向量量化，显著提升效率。实验表明，KBest在性能上超越运行在x86 CPU上的主流向量搜索库，优化使查询吞吐提升超过2倍。目前，KBest已服务于内部业务及外部企业客户，日均处理数千万次查询。
原文摘要 · Abstract (English)
Vector search, which returns the vectors most similar to a given query vector from a large vector dataset, underlies many important applications such as search, recommendation, and LLMs. To be economic, vector search needs to be efficient to reduce the resources required by a given query workload. However, existing vector search libraries (e.g., Faiss and DiskANN) are optimized for x86 CPU architectures (i.e., Intel and AMD CPUs) while Huawei Kunpeng CPUs are based on the ARM architecture and competitive in compute power. In this paper, we present KBest as a vector search library tailored for the latest Kunpeng 920 CPUs. To be efficient, KBest incorporates extensive hardware-aware and algorithmic optimizations, which include single-instruction-multiple-data (SIMD) accelerated distance computation, data prefetch, index refinement, early termination, and vector quantization. Experiment results show that KBest outperforms SOTA vector search libraries running on x86 CPUs, and our optimizations can improve the query throughput by over 2x. Currently, KBest serves applications from both our internal business and external enterprise clients with tens of millions of queries on a daily basis.
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。