原书出处
Data Management at Scale
04 Data Product Management
Managing only full snapshots is easy to implement. At each ETL cycle, the entire output is stored as an immutable snapshot. After that, table schemas can be replaced using the location in which the new data is stored. The container is a collection of all of these point-in-time snapshots. Some practitioners might argue that this approach results in data duplication. I don’t consider this a
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Data Management at Scale/_epub-src/OEBPS/ch04.html
EPUB SHA-256:48aa9e8e81be8a24243cd3a468bf103704d3a46e5c99edaa90b7433f7c941043