The GAP Benchmark Suite

Abstract

We present a graph processing benchmark suite with the goal of helping to standardize graph processing evaluations. Fewer differences between graph processing evaluations will make it easier to compare different research efforts and quantify improvements. The benchmark not only specifies graph kernels, input graphs, and evaluation methodologies, but it also provides optimized baseline implementations. These baseline implementations are representative of state-of-the-art performance, and thus new contributions should outperform them to demonstrate an improvement. The input graphs are sized appropriately for shared memory platforms, but any implementation on any platform that conforms to the benchmark's specifications could be compared. This benchmark suite can be used in a variety of settings. Graph framework developers can demonstrate the generality of their programming model by implementing all of the benchmark's kernels and delivering competitive performance on all of the benchmark's graphs. Algorithm designers can use the input graphs and the baseline implementations to demonstrate their contribution. Platform designers and performance analysts can use the suite as a workload representative of graph processing.

small revisions to correspond to v1.0

Cited by in corpus (24)

• Theoretically Efficient Parallel Graph Algorithms Can Be Fast and Scalable
• Graph Processing on FPGAs: Taxonomy, Survey, Challenges
• The gem5 Simulator: Version 20.0+
• A Probabilistic Machine Learning Approach to Scheduling Parallel Loops with Bayesian Optimization
• Delta-stepping SSSP: from Vertices and Edges to GraphBLAS Implementations
• Specifying and Testing GPU Workgroup Progress Models
• EMOGI: Efficient Memory-access for Out-of-memory Graph-traversal In GPUs
• ConnectIt: A Framework for Static and Incremental Parallel Graph Connectivity Algorithms
• ARC: DVFS-Aware Asymmetric-Retention STT-RAM Caches for Energy-Efficient Multicore Processors
• Die-Stacked DRAM: Memory, Cache, or MemCache?
• Log(Graph): A Near-Optimal High-Performance Graph Representation
• Load Driven Branch Predictor (LDBP)
• The LDBC Graphalytics Benchmark
• Pangolin: An Efficient and Flexible Graph Pattern Mining System on CPU and GPU
• Optimizing Ordered Graph Algorithms with GraphIt
• Sage: Parallel Semi-Asymmetric Graph Algorithms for NVRAMs
• SCART: Predicting STT-RAM Cache Retention Times Using Machine Learning
• Flexible Support for Fast Parallel Commutative Updates
• Efficient Stepping Algorithms and Implementations for Parallel Shortest Paths
• Benchmarking Graph Data Management and Processing Systems: A Survey
• Leveraging Architectural Support of Three Page Sizes with Trident
• Effective Cache Apportioning for Performance Isolation Under Compiler Guidance
• On Dynamic Precision Scaling
• ButterFly BFS -- An Efficient Communication Pattern for Multi Node Traversals