High-Throughput Graph Structures: Trie-Based Persistence with Adaptive Copying

• Elvis Mondal

  Author

This paper presents a novel framework for managing persistent graph data, focusing on achieving favorable performance characteristics. It describes a system designed with a special key-mapped trie as the primary data structure. The innovation is in the design of adaptive copying and chunking to greatly reduce memory overhead and enhance the speed of different graph operations. This paper explores the architectural benefits of such persistent designs and demonstrates their efficiency and scalability on complex graph management problems that require maintaining multiple historical versions. According to the experimental results, our approach achieves a substantial reduction in memory usage over traditional persistent graph structures, while also exhibiting competitive operation performance across a wide range of workloads.

• PDF

Copyright (c) 2026 International Journal of Intelligent Systems and Data Science

This work is licensed under a Creative Commons Attribution 4.0 International License.