KBest: Efficient Vector Search on Kunpeng CPU

• URL: http://arxiv.org/abs/2508.03016v2
• Date: Thu, 07 Aug 2025 03:48:34 GMT
• Title: KBest: Efficient Vector Search on Kunpeng CPU
• Authors: Kaihao Ma, Meiling Wang, Senkevich Oleg, Zijian Li, Daihao Xue, Dmitriy Malyshev, Yangming Lv, Shihai Xiao, Xiao Yan, Radionov Alexander, Weidi Zeng, Yuanzhan Gao, Zhiyu Zou, Xin Yao, Lin Liu, Junhao Wu, Yiding Liu, Yaoyao Fu, Gongyi Wang, Gong Zhang, Fei Yi, Yingfan Liu,
• Abstract summary: KBest is a vector search library tailored for the latest Huawei Kunpeng 920 CPUs.<n>To be efficient, KBest incorporates extensive hardware-aware and algorithmic optimizations.<n>Experiment results show that KBest outperforms SOTA vector search libraries running on x86 CPUs.
• Score: 21.419014075922657
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• NGPU-LM: GPU-Accelerated N-Gram Language Model for Context-Biasing in Greedy ASR Decoding [54.88765757043535]
  This work rethinks data structures for statistical n-gram language models to enable fast and parallel operations for GPU-optimized inference.<n>Our approach, named NGPU-LM, introduces customizable greedy decoding for all major ASR model types with less than 7% computational overhead.<n>The proposed approach can eliminate more than 50% of the accuracy gap between greedy and beam search for out-of-domain scenarios while avoiding significant slowdown caused by beam search.
  arXiv  Detail & Related papers  (2025-05-28T20:43:10Z)
• Bang for the Buck: Vector Search on Cloud CPUs [0.0]
  We show that CPU microarchitectures available in the cloud perform significantly differently across vector search scenarios.<n>For instance, in an IVF index on float32 vectors, AMD's Zen4 gives almost 3x more queries per second (QPS) compared to Intel's Sapphire Rapids.<n>We hope to guide users in getting the best "bang for the buck" when deploying vector search systems.
  arXiv  Detail & Related papers  (2025-05-12T14:44:21Z)
• RetrievalAttention: Accelerating Long-Context LLM Inference via Vector Retrieval [24.472784635757016]
  RetrievalAttention is a training-free approach to both accelerate attention computation and reduce GPU memory consumption.<n>We show that RetrievalAttention achieves near full attention accuracy while only requiring access to 1--3% of the data.
  arXiv  Detail & Related papers  (2024-09-16T17:59:52Z)
• LLM-Vectorizer: LLM-based Verified Loop Vectorizer [12.048697450464935]
  Large-language models (LLMs) can generate vectorized code from scalar programs that process individual array elements. LLMs are capable of producing high performance vectorized code with run-time speedup ranging from 1.1x to 9.4x. Our approach is able to verify 38.2% of vectorizations as correct on the TSVC benchmark dataset.
  arXiv  Detail & Related papers  (2024-06-07T07:04:26Z)
• Locally-Adaptive Quantization for Streaming Vector Search [1.151101202055732]
  Locally-Adaptive Vector Quantization (LVQ), a highly efficient vector compression method, yields state-of-the-art search performance for non-evolving databases. We introduce two improvements of LVQ: Turbo LVQ and multi-means LVQ that boost its search performance by up to 28% and 27%. Our studies show that LVQ and its new variants enable blazing fast vector search, outperforming its closest competitor by up to 9.4x for identically distributed data.
  arXiv  Detail & Related papers  (2024-02-03T05:43:39Z)
• The Faiss library [54.589857872477445]
  Faiss is a toolkit of indexing methods and related primitives used to search, cluster, compress and transform vectors.<n>This paper describes the trade-off space of vector search and the design principles of Faiss in terms of structure, approach to optimization and interfacing.
  arXiv  Detail & Related papers  (2024-01-16T11:12:36Z)
• INR-Arch: A Dataflow Architecture and Compiler for Arbitrary-Order Gradient Computations in Implicit Neural Representation Processing [66.00729477511219]
  Given a function represented as a computation graph, traditional architectures face challenges in efficiently computing its nth-order gradient. We introduce INR-Arch, a framework that transforms the computation graph of an nth-order gradient into a hardware-optimized dataflow architecture. We present results that demonstrate 1.8-4.8x and 1.5-3.6x speedup compared to CPU and GPU baselines respectively.
  arXiv  Detail & Related papers  (2023-08-11T04:24:39Z)
• Harnessing Deep Learning and HPC Kernels via High-Level Loop and Tensor Abstractions on CPU Architectures [67.47328776279204]
  This work introduces a framework to develop efficient, portable Deep Learning and High Performance Computing kernels. We decompose the kernel development in two steps: 1) Expressing the computational core using Processing Primitives (TPPs) and 2) Expressing the logical loops around TPPs in a high-level, declarative fashion. We demonstrate the efficacy of our approach using standalone kernels and end-to-end workloads that outperform state-of-the-art implementations on diverse CPU platforms.
  arXiv  Detail & Related papers  (2023-04-25T05:04:44Z)
• CITADEL: Conditional Token Interaction via Dynamic Lexical Routing for Efficient and Effective Multi-Vector Retrieval [72.90850213615427]
  Multi-vector retrieval methods combine the merits of sparse (e.g. BM25) and dense (e.g. DPR) retrievers. These methods are orders of magnitude slower and need much more space to store their indices compared to their single-vector counterparts. We propose conditional token interaction via dynamic lexical routing, namely CITADEL, for efficient and effective multi-vector retrieval.
  arXiv  Detail & Related papers  (2022-11-18T18:27:35Z)
• IRLI: Iterative Re-partitioning for Learning to Index [104.72641345738425]
  Methods have to trade between obtaining high accuracy while maintaining load balance and scalability in distributed settings. We propose a novel approach called IRLI, which iteratively partitions the items by learning the relevant buckets directly from the query-item relevance data. We mathematically show that IRLI retrieves the correct item with high probability under very natural assumptions and provides superior load balancing.
  arXiv  Detail & Related papers  (2021-03-17T23:13:25Z)
• Accelerating SLIDE Deep Learning on Modern CPUs: Vectorization, Quantizations, Memory Optimizations, and More [26.748770505062378]
  SLIDE is a C++ implementation of a sparse hash table based back-propagation. We show how SLIDE's computations allow for a unique possibility of vectorization via AVX (Advanced Vector Extensions-512) Our experiments are focused on large (hundreds of millions of parameters) recommendation and NLP models.
  arXiv  Detail & Related papers  (2021-03-06T02:13:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.