Towards Polyglot Data Processing in Social Networks using the Hadoop-Spark ecosystem
Antony Seabra, Sergio Lifschitz
TL;DR
This paper tackles scalable social network data processing by advocating a polyglot data processing approach within the Hadoop-Spark ecosystem, leveraging Hive, HBase, and GraphX to tailor compute and storage to specific analytics tasks. It defines three representative Twitter-driven tasks—identifying influential users, extracting dominant terms, and mapping user relationships—and evaluates them on a 3-node Google Dataproc cluster using ingestion into HDFS and storage across Hive and HBase. The results show task-level performance differences across engines (Hive vs HBase; MapReduce vs Spark; GraphFrames) and indicate favorable scalability as dataset size grows, with ongoing benefits from a polyglot configuration. The work highlights practical implications for designing flexible, efficient pipelines for social network analytics and outlines directions for future enhancements, including real-time processing and integration with emerging tools like Kafka, Flink, and Hudi.
Abstract
This article explores the use of the Hadoop-Spark ecosystem for social media data processing, adopting a polyglot approach with the integration of various computation and storage technologies, such as Hive, HBase and GraphX. We discuss specific tasks involved in processing social network data, such as calculating user influence, counting the most frequent terms in messages and identifying social relationships among users and groups. We conducted a series of empirical performance assessments, focusing on executing selected tasks and measuring their execution time within the Hadoop-Spark cluster. These insights offer a detailed quantitative analysis of the performance efficiency of the ecosystem tools. We conclude by highlighting the potential of the Hadoop-Spark ecosystem tools for advancing research in social networks and related fields.