Agrios: A hybrid approach to big array analytics

Abstract — Hybrid systems for analyzing big data integrate an analytic tool and a dedicated data-management platform. The necessary movement of data between the components of a hybrid system can lead to performance problems, if that movement is not managed effectively. We present Agrios, a hybrid analytic system for array-structured data, integrating R and SciDB. Agrios minimizes data movement between the two components of the hybrid, using techniques repurposed from relational database query optimization. Keywords—big-array analytics; query optimization; R; SciDB; I