Taming Hybrid-Cloud Fast and Scalable Graph Analytics at Twitter

We have witnessed a boosted demand for graph analytics at Twitter in recent years, and graph analytics has become one of the key parts of Twitter's large-scale data analytics and machine learning for driving engagement, serving the most relevant content, and promoting healthier conversations. However, infrastructure for graph analytics has historically not been an area of investment at Twitter, resulting in a long timeline and huge engineering effort for each project to deal with graphs at the Twitter scale. How do we build a unified graph analytics user experience to fulfill modern data analytics on various graph scales spanning from thousands to hundreds of billions of vertices and edges? To bring fast and scalable graph analytics capability into production, we investigate the challenges we are facing in largescale graph analytics at Twitter and propose a unified graph analytics platform for efficient, scalable, and reliable graph analytics across on-premises and cloud to fulfill the requirements of diverse graph use cases and challenging scales. We also conduct quantitative benchmarking on Twitter's production-level graph use cases between popular graph analytics frameworks to certify our solution.

I. Introduction

As one of the largest social media globally, Twitter manipulates multiple large-scale graphs with billions of vertices and edges. Implementing graph analytics on such large-scale graphs has been a long-lasting pain. Moreover, we have witnessed a growing need for large-scale graph analytics at Twitter in recent years. For example, recommendation teams are running PageRank and topic similarity tasks on the user-follow graph to measure the influence of Twitter users. Health teams are running multi-account detection on the safety graph to detect any or all Twitter accounts owned by the same person and combined connected users jobs to build relationships between users.