Dynamic graph processing is becoming increasingly critical across a wide range of domains, including social networks, financial transactions, and business intelligence. Its effectiveness relies heavily on optimizations in both storage and analytics, which are essential for improving system performance, throughput, and scalability. While dynamic graph processing has attracted significant research attention and yielded notable progress, a comprehensive analysis that integrates advancements in both dynamic graph storage and analytics remains lacking. To address this gap, this paper presents a thorough review of stateof-the-art techniques that support dynamic graph processing, with a particular focus on storage and analytical methods. Specifically, we first outline the fundamental challenges and core design principles in the field. Then, we systematically classify and summarize existing approaches, encompassing dynamic graph storage and analytics optimizations across both CPU and GPU platforms. Finally, we identify key research gaps and suggest promising directions for future work. This survey presents a comprehensive and up-to-date review of the literature on dynamic graph processing, offering valuable insights for both new and established researchers and contributing to the advancement of the field. The related materials for this paper are available at:https://github.com/yzhang610/DynGraphSurvey.

Dynamic graph storage and analytics have garnered increasing attention and found widespread application across various domains. However, real-world graphs own the characteristics of an inherently sparse, dynamic, and irregular nature, presenting significant challenges for efficient processing. These challenges encompass limited update throughput, suboptimal analytics performance, and difficulty in effectively supporting diverse and evolving workloads. To address these challenges, we design and implement TripleGraph, a high-throughput data structure for dynamic graphs that supports diverse workloads. Specifically, a hash-based bucket array is proposed to store vertices, enabling efficient vertex indexing. Then, we propose a diverse-workload-friendly two-layer edge storage strategy (dynamically mutable short and sorted arrays) that supports highthroughput graph updates while facilitating efficient graph analytics and pattern-matching. Besides, in terms of concurrency control, we introduce a fine-grained optimistic locking coupling scheme and an adaptive optimistic locking conversion mechanism to ensure data consistency and enhance system throughput. Extensive experimental results demonstrate that TripleGraph achieves average speedups of$23.97 \times 1.93 \times, 3.52 \times$, and 3.07 × over STINGER, GraphOne, Teseo, and Sortledton, respectively, on graph update workloads. Moreover, TripleGraph consistently outperforms STINGER, GraphOne, and Teseo in graph analytics and pattern-matching workloads.

Recently, large-scale graph processing on CPU-GPU heterogeneous platforms has attracted considerable attention. However, disparities in memory bandwidth and parallel computational capabilities between CPUs and GPUs, coupled with the irregular structure of graphs and the inherent unpredictability of graph algorithms, often lead to inefficient utilization of CPU-GPU hardware resources, ultimately degrading graph processing performance. To address this, we propose and implement CoDgraph, a co-design framework for high-performance graph processing on CPU-GPU heterogeneous platforms. Specifically, we introduce a fine-grained partitioning strategy to balance workloads, minimize communication overhead, and enhance data locality. Next, we develop an adaptive co-scheduling computing scheme, leveraging a cost model that accounts for CPU and GPU hardware resources to improve system utilization. Finally, to further optimize largescale graph processing, we design and implement an efficient overlapping pipeline execution mode that employs asynchronous parallel execution. Extensive evaluations demonstrate that CoDgraph outperforms state-of-the-art CPU and CPU-GPU graph processing systems, including Ligra (CoDgraph is$14.59 \times$faster on average) and Subway (CoDgraph is$4.17 \times$faster on average). In addition, CoDgraph also has comparable performance to the advanced in-memory graph processing Tigr on GPU and shows good scalability for different graph scales and CPU-GPU heterogeneous platforms.

Masked Graph Autoencoders (MGAEs), represented by GraphMAE and GraphMAE2, which utilize masked feature (or structure) reconstruction strategies, have demonstrated the potential to surpass contrastive learning. However, current masked reconstruction strategies primarily rely on random strategies, only prove effective on reliable graph data. Therefore, these popular methods face immediate robustness deficiencies issues. Firstly, when the graph is unreliable or under adversarial attacks, the selection of nodes for masked reconstruction has a significant impact on downstream tasks. Secondly, the reconstructed features contains redundant components. In this paper, to overcome the non-robustness caused by randomness, we provide a theoretical analysis and evaluation of the robustness of state-of-the-art MGAEs. Additionally, we design two lightweight plug-and-play tools: Box-Based Weighted Reliability Ranking Masking Strategy and Decoupled Feature Reconstruction. Without incurring additional time overhead, these tools provide a defense armor against adversarial attacks for MGAEs, significantly boosting the robustness performance of downstream tasks. Extensive experiments on real-world graphs attacked by various attacks demonstrate our designs have a considerable robust expressive ability. Especially on datasets with large perturbations, the defense performance could even be improved by up to 20%.

With the continuous growth of user scale and application data, the demand for large-scale concurrent graph processing is increasing. Typically, large-scale concurrent graph processing jobs need to process corresponding snapshots of dynamically changing graph data to obtain information at different time points. To enhance the throughput of such applications, current solutions concurrently process multiple graph snapshots on the GPU. However, when dealing with rapidly changing graph data, transferring multiple snapshots of concurrent jobs to the GPU results in high data transfer overhead between CPU and GPU. Additionally, the execution mode of existing work suffers from underutilization of GPU computational resources. In this work, we introduce CGCGraph, which can be integrated into existing GPU graph processing systems like Subway, to enable efficient concurrent graph snapshot processing jobs and enhance overall system resource utilization. The key idea is to offload unshared graph data of multiple concurrent snapshots to the CPU, reducing CPU-GPU transfer overhead. By implementing CPU-GPU co-execution, there is potential for enhanced utilization of GPU computing resources. Specifically, CGCGraph leverages kernel fusion to process shared graph data concurrently on the GPU, while executing all snapshots in parallel on the CPU, with each snapshot assigned a dedicated thread. This approach enables efficient concurrent processing within a novel CPU-GPU co-execution model, incorporating three optimization strategies targeting storage, computation, and synchronization. We integrate CGCGraph with Subway, an existing system designed for out-of-GPU-memory static graph processing. Experimental results show that the integration of CGCGraph with current GPU-based systems obtains performance improvements ranging from 1.7 to 4.5 times.

Recently, Message Passing Neural Networks (MPNNs) have become significant popular frameworks in graph neural networks (GNNs) for solve the graph representation learning(GRL). However, MPNNs overlook the importance of graph topology information and make it challenging to effectively exchange information between nodes with similar structure. To address this issue, we propose a novel model, that serves as an optimization technique being compatible with almost every MPNN model. Our method captures both local and global structural information simultaneously. Additionally, we adopt a topology-aware graph to integrate the local and global structural information into MPNNs. Subsequently, we introduce a model named Structure-Aware Graph Representation Learning (SAGRL), that can capture and exchange graph structural information between nodes with similar structures. We demonstrate the result of our method separately on node classification and graph classification tasks, validating the effectiveness of our approach. Furthermore, we employ visualization and ablation experiments to further validate our method.