Real-world graphs are often dynamic and evolve over time. It is crucial for storing and querying a graph’s evolution in graph databases. However, existing works either suffer from high storage overhead or lack efficient temporal query support, or both. In this paper, we propose \(\textsf {AeonG}\) , a new graph database with built-in temporal support. Based on a novel temporal graph model, we build \(\textsf {AeonG}\) with a hybrid storage engine and an efficient temporal query engine. The storage engine consists of current storage to manage the most recent versions of graph objects, and historical storage to manage previous versions. This separation minimizes the performance degradation when querying the most recent graph object versions. To reduce the historical storage overhead, we propose an anchor+delta strategy, in which we periodically create a complete version (namely anchor) of a graph object, and maintain every change (namely delta) between two adjacent anchors of the same object. In the query engine, we propose an anchor-based version retrieval technique to skip unnecessary historical version traversals to boost temporal query processing. Further, we extend \(\textsf {AeonG}\) into a cloud-native database with disaggregated compute and storage layers, thus enabling elastic and scalable management of temporal graph data. Extensive experiments are conducted on both real and synthetic datasets. The results show that \(\textsf {AeonG}\) achieves up to 5.73 \(\times \) lower storage consumption and 2.57 \(\times \) lower temporal query latency against state-of-the-art approaches, while introducing only 9.74% performance degradation for supporting temporal features.
