SQL databases use two entirely different group by algorithms. The first one, the hash algorithm, aggregates the input records in a temporary hash table. Once all input records are processed, the hash table is returned as the result. The second algorithm, the sort/group algorithm, first sorts the input data by the grouping key so that the rows of each group follow each other in immediate succession. Afterwards, the database just needs to aggregate them. In general, both algorithms need to materialize an intermediate state, so they are not executed in a pipelined manner. Nevertheless the sort/group algorithm can use an index to avoid the sort operation, thus enabling a pipelined group by.

 

 

Consider the following query. It delivers yesterday's revenue grouped by 

PRODUCT_ID:

SELECT product_id, sum(eur_value)  FROM sales WHERE sale_date = TRUNC(sysdate) - INTERVAL '1' DAY GROUP BY product_id

Knowing the index on SALE_DATE and PRODUCT_ID from the , the sort/group algorithm is more appropriate because an INDEX RANGE SCAN automatically delivers the rows in the required order. That means the database avoids materialization because it does not need an explicit sort operation—the group by is executed in a pipelined manner.

oracle:---------------------------------------------------------------|Id |Operation                    | Name        | Rows | Cost |---------------------------------------------------------------| 0 |SELECT STATEMENT             |             |   17 |  192 || 1 | SORT GROUP BY NOSORT        |             |   17 |  192 || 2 |  TABLE ACCESS BY INDEX ROWID| SALES       |  321 |  192 ||*3 |   INDEX RANGE SCAN          | SALES_DT_PR |  321 |    3 |---------------------------------------------------------------

 

The Oracle database's execution plan marks a pipelined 

SORT GROUP BY operation with the 

NOSORT addendum. The execution plan of other databases does not mention any sort operation at all. 

 

The pipelined 

group by has the same prerequisites as the pipelined 

order by, except there are no 

ASC and 

DESC modifiers. That means that defining an index with 

ASC/

DESC modifiers should not affect pipelined 

group by execution. The same is true for 

NULLS FIRST/

LAST. Nevertheless there are databases that cannot properly use an 

ASC/

DESC index for a pipelined 

group by.

 

For PostgreSQL, you must add an order by clause to make an index with NULLS LAST sorting usable for a pipelined group by. The Oracle database cannot read an index backwards in order to execute a pipelined group by that is followed by an order by. More details are available in the respective appendices: , .

 

If we extend the query to consider all sales since yesterday, as we did in the example for the pipelined order by, it prevents the pipelined group by for the same reason as before: the INDEX RANGE SCAN does not deliver the rows ordered by the grouping key.

SELECT product_id, sum(eur_value)  FROM sales WHERE sale_date >= TRUNC(sysdate) - INTERVAL '1' DAY GROUP BY product_id

 

Oracle:---------------------------------------------------------------|Id |Operation                    | Name        | Rows | Cost |---------------------------------------------------------------| 0 |SELECT STATEMENT             |             |   24 |  356 || 1 | HASH GROUP BY               |             |   24 |  356 || 2 |  TABLE ACCESS BY INDEX ROWID| SALES       |  596 |  355 ||*3 |   INDEX RANGE SCAN          | SALES_DT_PR |  596 |    4 |---------------------------------------------------------------

 

Instead, the Oracle database uses the hash algorithm. The advantage of the hash algorithm is that it only needs to buffer the aggregated result, whereas the sort/group algorithm materializes the complete input set. In other words: the hash algorithm needs less memory.

As with pipelined order by, a fast execution is not the most important aspect of the pipelined group by execution. It is more important that the database executes it in a pipelined manner and delivers the first result before reading the entire input. 

 

参考：

http://use-the-index-luke.com/sql/sorting-grouping/indexed-group-by