1. 将 aggregation 操作下压，获得部分的聚合结果。
  2. 进行汇总的方式获得最终结果。
  2. *************************** 1. row ***************************
  3. Query Plan: 
  4. =============================================
  5. ---------------------------------------------
  6. |0 |HASH GROUP BY       |    |1000     |3395|
  8. |2 |  EXCHANGE OUT DISTR|    |1000     |2740|
  9. |3 |   HASH GROUP BY    |    |1000     |2740|
  10. |4 |    TABLE SCAN      |t2  |4000     |499 |
  11. =============================================
  14.   0 - output([T_FUN_SUM(T_FUN_SUM(t2.c1))]), filter(nil), 
  15.       group([t2.c2]), agg_func([T_FUN_SUM(T_FUN_SUM(t2.c1))])
  16.   1 - output([T_FUN_SUM(t2.c1)], [t2.c2]), filter(nil)
  17.   2 - output([T_FUN_SUM(t2.c1)], [t2.c2]), filter(nil)
  18.   3 - output([T_FUN_SUM(t2.c1)], [t2.c2]), filter(nil), 
  19.       group([t2.c2]), agg_func([T_FUN_SUM(t2.c1)])