Awesome-Graph-Neural-Networks

This is the repo to collect latest materials of GNN, mainly focus on system context. Welcome to contribute!

System Designs

GNN System Design (Need to be further organized)

Distributed Sys Design

• [KDD'22] Distributed Hybrid CPU and GPU training for Graph Neural Networks on Billion-Scale Heterogeneous Graphs [Paper]
• [MLSys'22] Sequential aggregation and rematerialization: Distributed full-batch training of graph neural networks on large graphs [Paper]
• [KDD'21] Global Neighbor Sampling for Mixed CPU-GPU Training onGiant Graphs [Paper]
• [OSDI'21] P3: Distributed Deep Graph Learning at Scale [Paper]
• [OSDI'21] Dorylus: Affordable, Scalable, and Accurate GNN Training with Distributed CPU Servers and Serverless Threads [Paper]
• [NSDI'21] GAIA: A System for Interactive Analysis on Distributed Graphs Using a High-Level Language [Paper]
• [EuroSys'21] DGCL: An Efficient Communication Library for Distributed GNN Training [Paper]
• [EuroSys'21] FlexGraph: a flexible and efficient distributed framework for GNN training [Paper]
• [IA3'20] Distdgl: distributed graph neural network training for billion-scale graphs [Paper]
• [SOSP'19] Knightking: a fast distributed graph random walk engine [Paper]

Single-machine Sys Design

• [VLDB'22] Marius++: Large-Scale Training of Graph Neural Networks on aSingle Machine [Paper] [Website] [Repo]
• [MLSys'22] Accelerating training and inference of graph neural networks with fast sampling and pipelining [Paper]
• [PPoPP'22] Accelerating Quantized Graph Neural Networks via GPU Tensor Core [Papers]
• [PPoPP'21] Understanding and Bridging the Gaps in Current GNN Performance Optimizations [Paper]
• [OSDI'21] GNNAdvisor: An Adaptive and Efficient Runtime System for GNN Acceleration on GPUs [Paper] [Presentation] [Repo] [Chinese Blog from ZobinHuang]
• [OSDI'21] Marius: Learning Massive Graph Embeddings on a Single Machine [Paper] [Website] [Repo]
• [VLDB'21] Large Graph Convolutional Network Training with GPU-Oriented Data Communication Architecture [Paper]
• [EuroSys'21] Accelerating graph sampling for graph machine learning using GPUs [Paper] [Repo]
• [EuroSys'21] Seastar: vertex-centric programming for graph neural networks [Paper]
• [MLSys'20] Improving the Accuracy, Scalability, and Performance of Graph Neural Networks with Roc [Paper]
• [VLDB'20] G3: When Graph Neural Networks Meet Parallel GraphProcessing Systems on GPUs [Paper]
• [MLSys'19] Optimizing DNN computation with relaxed graph substitutions [Paper]
• [ATC'19] NeuGraph: Parallel Deep Neural Network Computation on Large Graphs [Paper] [Presentation] [Chinese Blog from ZobinHuang]
• [WWW'19] GraphVite: A High-Performance CPU-GPU Hybrid System forNode Embedding [Paper]
• [Unknown] PyTorch-Direct: Enabling GPU Centric Data Access for Very Large Graph Neural Network Training with Irregular Accesses [Paper]

Traditional Graph Processing System Design

Graph Analyse

• [PPoPP'22] Scaling Graph Traversal to 281 Trillion Edges with 40 Million Cores [Paper]
• [EuroSys'21] Tripoline: generalized incremental graph processing via graph triangle inequality [Paper]
• [EuroSys'21] DZiG: Sparsity-Aware Incremental Processing of Streaming Graphs [Paper]
• [EuroSys'20] Subway: Minimizing Data Transfer duringOut-of-GPU-Memory Graph Processing [Paper]
• [ATC'19] SIMD-X: Programming and Processing of Graph Algorithms on GPUs [Paper] [Repo]
• [ASPLOS'18] Tigr: Transforming Irregular Graphs forGPU-Friendly Graph Processing [Paper] [Repo]
• [PPoPP'16] Gunrock: A High-Performance Graph Processing Library on the GPU [Paper] [Repo]
• [SC'15] Enterprise: Breadth-First Graph Traversal on GPUs [Paper] [Repo]
• [HPDC'14] Cusha: vertex-centric graph pro-cessing on gpus [Paper] [Repo]
• [OSDI'12] PowerGraph: Distributed Graph-Parallel Computation on Natural Graphs [Paper]
• [SIGMOD'10] Pregel: A System for Large-Scale Graph Processing [Paper]

Graph Mining

• [EuroSys'21] Tesseract: distributed, general graph pattern mining on evolving graphs [Paper]
• [VLDB'20] Pangolin: An Efficient and Flexible Graph Mining Systemon CPU and GPU [Paper]
• [SOSP'19] AutoMine: Harmonizing High-Level Abstraction and High Performance for Graph Mining [Paper]
• [OSDI'18] ASAP: Fast, Approximate Graph Pattern Mining at Scale [Paper]
• [SOSP'15] Arabesque: A System for Distributed Graph Mining [Paper]

Open Source System

• Deep Graph Library (DGL) [Website] [Paper] [Repo]
• PyG (PyTorch Geometric) [Repo] [Paper]
• Cogdl [Repo] [Paper]
• Graph-Learn [Repo] [Paper]
• PyTorch-BigGraph [Repo] [Paper]
• Gunrock [Repo] [Paper]
• SNAP (Stanford Network Analysis Platform) [Website] [Repo]

Open Source Dataset

• Open Graph Benchmark [Website]