Is your feature request related to a problem or challenge?
In the above EPIC's aggregate refactor, it's now using a variant of AggregateHashTable<ParitalSkip> to handle the convert_to_states() optimization (see #22899 for this optimization)
https://github.com/apache/datafusion/blob/main/datafusion/physical-plan/src/aggregates/hash_table.rs#L601
The only needed fields are only GroupsAccumulators, and we don't have to carry along the heavy hash table, so we can simplify AggregateHashTable<PartialSkip> into a simpler struct.
Describe the solution you'd like
No response
Describe alternatives you've considered
No response
Additional context
No response
Is your feature request related to a problem or challenge?
In the above EPIC's aggregate refactor, it's now using a variant of
AggregateHashTable<ParitalSkip>to handle theconvert_to_states()optimization (see #22899 for this optimization)https://github.com/apache/datafusion/blob/main/datafusion/physical-plan/src/aggregates/hash_table.rs#L601
The only needed fields are only
GroupsAccumulators, and we don't have to carry along the heavy hash table, so we can simplifyAggregateHashTable<PartialSkip>into a simpler struct.Describe the solution you'd like
No response
Describe alternatives you've considered
No response
Additional context
No response