关键词：大规模图数据；处理；技术指南
摘 要：With emphasis on Apache Giraph and the GraphLab framework, this article introduces and compares open source solutions for processing large volumes of graph data. The growth of graph-structured data in modern applications such as social networks and knowledge bases creates a crucial need for scalable platforms and parallel architectures that can process it in bulk. Despite its prominent role in big data analytics, MapReduce is not the optimal programming model for graph processing. This article explains why and then explores systems in development to tackle the graph-processing challenge.