Distance Comparison Operations Are Not Silver Bullets in Vector Similarity Search: A Benchmark Study on Their Merits and Limits
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908934714949632 |
|---|---|
| author | Zheng, Zhuanglin Zeng, Yuxiang Liu, Chenchen Chi, Yunzhen Yang, Binhan Tong, Yongxin |
| author_facet | Zheng, Zhuanglin Zeng, Yuxiang Liu, Chenchen Chi, Yunzhen Yang, Binhan Tong, Yongxin |
| contents | Distance Comparison Operations (DCOs), which decide whether the distance between a data vector and a query is within a threshold, are a critical performance bottleneck in vector similarity search. Recent DCO methods that avoid full-dimensional distance computations promise significant speedups, but their readiness for production vector database systems remains an open question. To address this, we conduct a comprehensive benchmark of 8 DCO algorithms across 10 datasets (with up to 100M vectors and 12,288 dimensions) and diverse hardware configurations (CPUs with/without SIMD, and GPUs). Our study reveals that these methods are not silver bullets: their efficiency is highly sensitive to data dimensionality, degrades under out-of-distribution queries, and is unstable across hardware. Yet, our evaluation also demonstrates often-overlooked merits: they can accelerate index construction and data updates. Despite these benefits, their unstable performance, which can be slower than a full-dimensional scan, leads us to conclude that recent algorithmic advancements in DCO are not yet ready for production deployment. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_02801 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Distance Comparison Operations Are Not Silver Bullets in Vector Similarity Search: A Benchmark Study on Their Merits and Limits Zheng, Zhuanglin Zeng, Yuxiang Liu, Chenchen Chi, Yunzhen Yang, Binhan Tong, Yongxin Databases Distance Comparison Operations (DCOs), which decide whether the distance between a data vector and a query is within a threshold, are a critical performance bottleneck in vector similarity search. Recent DCO methods that avoid full-dimensional distance computations promise significant speedups, but their readiness for production vector database systems remains an open question. To address this, we conduct a comprehensive benchmark of 8 DCO algorithms across 10 datasets (with up to 100M vectors and 12,288 dimensions) and diverse hardware configurations (CPUs with/without SIMD, and GPUs). Our study reveals that these methods are not silver bullets: their efficiency is highly sensitive to data dimensionality, degrades under out-of-distribution queries, and is unstable across hardware. Yet, our evaluation also demonstrates often-overlooked merits: they can accelerate index construction and data updates. Despite these benefits, their unstable performance, which can be slower than a full-dimensional scan, leads us to conclude that recent algorithmic advancements in DCO are not yet ready for production deployment. |
| title | Distance Comparison Operations Are Not Silver Bullets in Vector Similarity Search: A Benchmark Study on Their Merits and Limits |
| topic | Databases |
| url | https://arxiv.org/abs/2604.02801 |