SheetReader: Efficient Specialized Spreadsheet Parsing

Spreadsheets are widely used for data exploration. Since spreadsheet systems have limited capabilities, users often need to load spreadsheets other science environments perform advanced analytics. However, current approaches loading suffer from either high runtime or memory usage, which hinders exploration on commodity systems. To make practical systems, we introduce a novel parser that minimizes usage by tightly coupling decompression and parsing. Furthermore, reduce the runtime, optimized spreadsheet-specific parsing routines employ parallelism. evaluate our approach, implement prototypes Excel into R Python environments. Our evaluation shows approach is up 3× faster while consuming 40× less than state-of-the-art approaches. The source code available at https://github.com/fhenz/SheetReader-r.