TEngineDB-V: An OLAP-Native Vector Search System for Large-$k$ Workloads at Tencent

2026-08-01

AI Summary

This work addresses the inefficiency of existing vector retrieval systems in supporting large-scale analytical queries with high k (10³–10⁵), which suffer from limited result scalability, weak analytical capabilities, and severe read and compute amplification. To overcome these limitations, the paper elevates vector search to a first-class analytical primitive within an OLAP engine and introduces a suite of optimizations: segment-decoupled global vector indexing, direction-aware product quantization, hierarchical residual refinement, and index-aware query rewriting. These techniques are tightly integrated with native OLAP storage and distributed execution frameworks. The proposed approach achieves both high recall and analytical efficiency, demonstrating up to a 145× speedup over systems like StarRocks in billion-scale production environments, with a practical performance gain of 52×.

Abstract

Vector search systems are essential infrastructure for modern data-driven applications. Large-$k$ analytical vector search, which retrieves $k=10^3$--$10^5$ results for analytics (e.g., aggregation, filtering, joins), is increasingly important for emerging workloads, including LLM data management and advertising analysis at Tencent. Existing systems remain inadequate: specialized vector databases often cap $k$ (e.g., $k \leq 10^4$) to satisfy tail-latency constraints and offer limited analytical support, while OLAP systems typically embed per-segment vector indexes as black boxes, causing severe read/compute amplification and preventing native query optimization. This paper presents TEngineDB-V, an OLAP-native vector search system for large-$k$ workloads. TEngineDB-V makes vector search a first-class analytical primitive in Tencent's OLAP engine through a global segment-decoupled index materialized as relational tables, eliminating scatter-gather execution, reducing amplification, and enabling native storage optimizations. It decomposes IVFPQ-based search into relational operators, integrates OLAP optimizations, and introduces DPPQ, which combines direction-aware quantization with hierarchical residual refinement to improve recall while preserving relational efficiency. TEngineDB-V further incorporates index-aware query rewriting and a distributed-aware cost model for efficient distributed execution. Experiments show that TEngineDB-V achieves up to a $145\times$ speedup over competitive systems such as StarRocks, and up to a $52\times$ improvement in 10-billion-scale production deployments.

Problem

• large-k vector search
• OLAP
• vector database
• analytical workload
• read/compute amplification

Innovation

• OLAP-native
• large-k vector search
• segment-decoupled index
• DPPQ
• relational vector operators