flyovercountry·13 ปีที่แล้ว·discussThe file format for hadoop tools is a sequence file. For most queries this is the second slowest format, after text.Rcfile or parquet would be a more interesting benchmark.
flyovercountry·13 ปีที่แล้ว·discussHDFS now supports placement groups with one use case being columnar storage.
Rcfile or parquet would be a more interesting benchmark.