TVA: A Version-aware Temporal Graph Storage System for Real-time Analytics

Jul 16, 2026

Upload by Himbo

TVA is a version-aware temporal graph storage system built to enable efficient temporal query processing for real-time analytics. It addresses limitations of existing systems that either lack native temporal support or incur high latency on temporal queries. TVA uses a multi-version architecture separating version metadata from property data, tailored structures (temporal tables and enhanced hopscotch hashing) to reduce random I/O, and a version-skipping strategy that reuses temporal results across neighborhood scans. Experiments show up to 9.9× lower query latency and 2.2× lower storage overhead versus state-of-the-art systems.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

Introduction

Motivating temporal queries

Challenges in existing graph storage systems

TVA: A Version-aware Temporal Graph Storage System for Real-time Analytics Wenhao Li Renmin University of China [ruclwh@ruc.edu.cn](ruclwh@ruc.edu.cn) Zhanhao Zhao Renmin University of China [zhanhaozhao@ruc.edu.cn](zhanhaozhao@ruc.edu.cn) Jinhao Dong Renmin University of China [imonedjh@gmail.com](imonedjh@gmail.com) Jiamin Hou Zhejiang Univ. & RUC [jiaminhou@zju.edu.cn](jiaminhou@zju.edu.cn) Wei Lu Renmin University of China [lu-wei@ruc.edu.cn](lu-wei@ruc.edu.cn) Yunhai Wang Renmin University of China [cloudseawang@gmail.com](cloudseawang@gmail.com) Xiaoyong Du Renmin University of China [duyong@ruc.edu.cn](duyong@ruc.edu.cn) arXiv :2607 .00406v 1 [ cs .DB] 1 Jul 2026 ABSTRACT Analyzing temporal graphs can reveal valuable insights that are typically hidden in static graphs. Unfortunately, existing graph storage systems either lack native temporal support or suffer from high latency when querying temporal graphs. This paper presents TVA, anew temporal graph storage system designed for efficient temporal query processing. First, TVA introduces a specialized multi-version storage architecture that separates version metadata from actual data, i.e., the property values associated with different versions of vertices and edges. This architecture enables efficient version retrieval for a vertex or edge by quickly locating valid version metadata and directly dereferencing it to access the corresponding property values. Second, we design tailored data structures, namely the temporal table and enhanced hopscotch-based hashing, to compactly organize the version metadata of adjacent vertices and edges, thus reducing random I/O for metadata lookups during the neighborhood scan initiated from a vertex. Finally, to further accelerate neighborhood scans over multiple vertices, we propose a versionskipping strategy that reuses temporal information obtained from prior scans, thereby avoiding redundant metadata lookups across scans. Empirical evaluations demonstrate that TVA achieves up to 9. 9× lower temporal query latency and 2 . 2× lower storage overhead compared to state-of-the-art temporal graph storage systems. PVLDB Reference Format: Wenhao Li, Zhanhao Zhao, Jinhao Dong, Jiamin Hou, Wei Lu, Yunhai Wang, and Xiaoyong Du. TVA: A Version-aware Temporal Graph Storage System for Real-time Analytics. PVLDB, 19(10): 2536-2548, 2026 . doi:10.14778/3828612.3828613 PVLDB Artifact Availability: The source code, data, and/or other artifacts have been made available at [https://github.com/Sakuraaa0/TVA.git](https://github.com/Sakuraaa0/TVA.git). 1 INTRODUCTION A temporal graph, also known as a dynamic or time-varying graph, represents the evolution of relationships between entities, the entities themselves, or both over time [11, 30] . Unlike static graphs, Wei Lu is the corresponding author. This work is licensed under the Creative Commons BY-NC-ND 4.0 International License. Visit [https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[ ](https://creativecommons.org/licenses/by-nc-nd/4.0/ to view a copy of)[this license. For any use beyond those covered by this license](this license. For any use beyond those covered by this license), [obtain permission by](obtain permission by)[emailing info@vldb.org. Copyright](emailing info@vldb.org. Copyright) is held by the owner/author(s). Publication rights licensed to the VLDB Endowment. Proceedings of the VLDB Endowment, Vol. 19, No. 10 ISSN 2150-8097 . doi:10.14778/3828612.3828613 􀝐0 Ten Years 􀝐􀯡 One Minutes 􀝐 􀯡+1 Figure 1: An Example of Financial Temporal Graph. temporal graphs retain rich temporal information, which is crucial for a wide range of applications, such as fraud detection [3, 4, 34] and social network analysis [18, 21, 28] . Example 1. Figure 1 shows a typical temporal graph in the financial domain, where vertices represent entities, such as users,

FAQs

What problem does TVA target in temporal graph analytics?

TVA targets high latency and the lack of native temporal support in existing graph storage systems when processing temporal queries efficiently for real-time analytics.

How does TVA organize data to speed up temporal queries?

TVA introduces a multi-version storage architecture that separates version metadata from actual property values, enabling fast retrieval of the correct version for vertices and edges.

Which techniques does TVA use to reduce I/O during neighborhood scans?

TVA uses a temporal table and enhanced hopscotch-based hashing to compactly store adjacent version metadata, lowering random I/O during metadata lookups.

Jul 16, 2026

Upload by Himbo