How many blocks are read from Disk to Buffer Cache?

Hi,

I am trying to understand this.

1. Assuming that I have a table, say of 1000 used blocks (below HWM) and I use the following sql statement which uses a Full Table Scan, to get rows which are in about 100 blocks. Will the server process gets or copies all the 1000 blocks and puts in the DB Buffer cache or only the blocks (100 in this case) which has the candidate records/rows.

    Select * from sales where purchase_date between ' ' and ' '; (For this test scenario, please assume that I don't have an index on purchase_date column).

2. When a query which returns only one row and uses an index to get to the candidate row/block, will it get only the block which has the record to the DBBC or the number of blocks are defined by db_file_multiblock_read_count?

Will some one please help me understand this?

I would appreciate your time and help

Thanks
TG Received on Tue Nov 11 2003 - 18:56:20 CST