read by other session occurs when two users need access to the same block of data. The first user reads the data from disk and places it in the buffer cache. The second user has to wait for the first users operation to complete so they are placed in to waiting. This is when the read by other session wait occurs. Unfortunately this is one of those events we need to "catch in the act" to properly resolve.
Recommendation:
Use the following queries
(during user load) to identify the
hot block(s):
SELECT p1
"file#", p2 "block#", p3 "class#"
FROM v$session_wait WHERE event =
'read by other session';
Once a common block has been
identified use this query to
identify the hot object:
SELECT
relative_fno, owner, segment_name,
segment_type FROM dba_extents WHERE
file_id = &file AND &block BETWEEN
block_id AND block_id + blocks - 1;
The next step is to tune queries that interact with this object. Secondly increasing PCTFREE and allowing for less dense blocks may address the problem, depending on the real root cause however the best solution is SQL tuning. A third option is to move the table to a smaller tablespace, rather than the current 16k tablespace size, to reduce the amount of data in each block.
Note: PCTFREE is still required even with ASSM because it determines how rows will be packed into blocks, while freelists, pctused and pctincrease are ignored with ASSM because they are used for transaction/block management.
