That hybrid approach is super cool. Thanks for sharing that!
@yinchuanwang381
3 жыл бұрын
Regarding the hybrid approach, even with the ability of calcite, we can have the push-down plan. However after we put it to spark, it still depends on the implementation of spark data source. For example, in calcite elastic search adapter, we can have some aggregations pushed down to es. But the spark es data source does not support aggregation push-down. What should we do for this case?
@reapi_com
4 жыл бұрын
wonderful introduction
@pranavdeshpande4942
2 жыл бұрын
Are any details available on how you squashed the pieces related to separate data sources into "Spark Tables"?
@pp6278
3 жыл бұрын
You haven't changed a bit, Chris. You look as young as when you were in Coventry.
Пікірлер: 9