google bigquery - identify group by vs group each in advance -


is there way figure out in advance (not trial , error) whether specific query should use group or group each by? saw after cardinality of ~60-70% asked use group each by. hard predict generate sql.

the usage of 'each' doesn't depend on query, on data. there small number of unique values group expression? use group by. there lot? use group each by.

the best strategy use group until "over limits error".

to go deeper "why?", can @ dremel paper started all. group runs in mixers, while group each gets pushed shards.

for other insights, check jcondit's answers @ resources exceeded during query execution.


Comments

Popular posts from this blog

java - Jmockit String final length method mocking Issue -

asp.net - Razor Page Hosted on IIS 6 Fails Every Morning -

c++ - wxwidget compiling on windows command prompt -