{"title":"GHVSA:基于图的高维矢量搜索加速器","authors":"Wei Yuan, Huawen Liang, Xi Jin","doi":"10.1016/j.sysarc.2025.103514","DOIUrl":null,"url":null,"abstract":"<div><div>Graph-based high-dimensional vector search is the most common method for retrieval and achieves the best performance. However, implementing graph-based high-dimensional vector search in industrial applications requires reduced redundant computation, high accuracy, and low latency, which existing works fail to provide. We propose GHVSA, a graph-based vector search architecture that achieves low on-chip memory footprint and high energy efficiency by narrowing down the scope of vector search (about 0.5% of the entire dataset), avoiding redundant computation(more than 50% in the distance computation phase and more than 90% in the sorting phase) and data reuse. Extensive evaluations on three datasets show that GHVSA achieves an average of 609.07×, 53.6×, 5.16×, and 6.06× better energy efficiency than CPUs, GPUs, DF-GAS, and VStore, respectively.</div></div>","PeriodicalId":50027,"journal":{"name":"Journal of Systems Architecture","volume":"167 ","pages":"Article 103514"},"PeriodicalIF":3.7000,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"GHVSA: Graph-based high-dimensional vector search accelerator\",\"authors\":\"Wei Yuan, Huawen Liang, Xi Jin\",\"doi\":\"10.1016/j.sysarc.2025.103514\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Graph-based high-dimensional vector search is the most common method for retrieval and achieves the best performance. However, implementing graph-based high-dimensional vector search in industrial applications requires reduced redundant computation, high accuracy, and low latency, which existing works fail to provide. We propose GHVSA, a graph-based vector search architecture that achieves low on-chip memory footprint and high energy efficiency by narrowing down the scope of vector search (about 0.5% of the entire dataset), avoiding redundant computation(more than 50% in the distance computation phase and more than 90% in the sorting phase) and data reuse. Extensive evaluations on three datasets show that GHVSA achieves an average of 609.07×, 53.6×, 5.16×, and 6.06× better energy efficiency than CPUs, GPUs, DF-GAS, and VStore, respectively.</div></div>\",\"PeriodicalId\":50027,\"journal\":{\"name\":\"Journal of Systems Architecture\",\"volume\":\"167 \",\"pages\":\"Article 103514\"},\"PeriodicalIF\":3.7000,\"publicationDate\":\"2025-07-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Systems Architecture\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S1383762125001869\",\"RegionNum\":2,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Systems Architecture","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1383762125001869","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, HARDWARE & ARCHITECTURE","Score":null,"Total":0}
Graph-based high-dimensional vector search is the most common method for retrieval and achieves the best performance. However, implementing graph-based high-dimensional vector search in industrial applications requires reduced redundant computation, high accuracy, and low latency, which existing works fail to provide. We propose GHVSA, a graph-based vector search architecture that achieves low on-chip memory footprint and high energy efficiency by narrowing down the scope of vector search (about 0.5% of the entire dataset), avoiding redundant computation(more than 50% in the distance computation phase and more than 90% in the sorting phase) and data reuse. Extensive evaluations on three datasets show that GHVSA achieves an average of 609.07×, 53.6×, 5.16×, and 6.06× better energy efficiency than CPUs, GPUs, DF-GAS, and VStore, respectively.
期刊介绍:
The Journal of Systems Architecture: Embedded Software Design (JSA) is a journal covering all design and architectural aspects related to embedded systems and software. It ranges from the microarchitecture level via the system software level up to the application-specific architecture level. Aspects such as real-time systems, operating systems, FPGA programming, programming languages, communications (limited to analysis and the software stack), mobile systems, parallel and distributed architectures as well as additional subjects in the computer and system architecture area will fall within the scope of this journal. Technology will not be a main focus, but its use and relevance to particular designs will be. Case studies are welcome but must contribute more than just a design for a particular piece of software.
Design automation of such systems including methodologies, techniques and tools for their design as well as novel designs of software components fall within the scope of this journal. Novel applications that use embedded systems are also central in this journal. While hardware is not a part of this journal hardware/software co-design methods that consider interplay between software and hardware components with and emphasis on software are also relevant here.