Architecting big data: Challenges, studies and forecasts

Compared with traditional data warehouse applications, big data analytics are huge and complex. To design a favorable architecture for big data analytics, this paper lists some key features for big data analytics, summarizes current main implementation platforms(parallel databases, MapReduce, and hybrid architectures based on them), and points their pros and cons. Some current researches are also investigated, our work are introduced and some challenging research problems in the future are discussed.