Spreadsheets for stream processing with unbounded windows and partitions

摘要: Stream processing is a computational paradigm that allows the analysis of live data streams as they are produced. This paper describes programming model, based on enhancements to spreadsheets, enables users with limited experience participate directly in development complex streaming applications. The model augments conventional spreadsheet features permit operating over unbounded sets despite finite interface provided by spreadsheet. new constructs include time-based windows and partitioning. We introduce compiler generates C++ code achieve integration existing stream systems. Our experimental study illustrates expressivity finds our implementation between 8x slower 2x faster than hand-written programs.