one sid process has over 20 sql_id associated with it and it is very slow
OS: HPDB: 11.1.0.7
we have an insert process sid: 639, it basically insert to a transaction table. and people complain the transactions piled up due to slowness.
I literally watched it by my eyes on toad and found out there are over around 20 sql_id associated with it.
and it has select then insert contantly. all explain plan seems ok.
How do I tune it from here?
Thanks in advance.
P.S. the wait event for all those sql_id are 'db file sequential read'.
SGA size is 9g, and pga size is 4g. This company dont believe in increase memory . So I would like to see what I can do to tune it.
0