Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> RH Linux 8 vs Windows XP -- Question...
Gentlemen...
I'm wondering if anybody has encountered something similar... I'm running Tom Kyte's script to create BIG_TABLE on two play databases I run, one on RH Linux 8, one on XP. They have been configured similarly, init.ora parameters are pretty much the same. Yet, look at these tkprof snippets:
XP
INSERT /*+ APPEND */ into big_table
select rownum+:b1, OWNER, OBJECT_NAME, SUBOBJECT_NAME, OBJECT_ID, DATA_OBJECT_ID, OBJECT_TYPE, CREATED, LAST_DDL_TIME, TIMESTAMP, STATUS, TEMPORARY, GENERATED, SECONDARY from big_table where rownum <= :b2-:b1 call count cpu elapsed disk query current
rows
------- ------ -------- ---------- ---------- ---------- ----------
Parse 1 0.00 0.00 0 0 0 0 Execute 6 5.37 10.97 8365 14635 2133 970885 Fetch 0 0.00 0.00 0 0 0 0
versus Linux:
call count cpu elapsed disk query current
rows
------- ------ -------- ---------- ---------- ---------- ----------
Parse 1 0.00 0.00 0 0 0 0 Execute 6 12.81 33.79 7601 14635 2130 970482 Fetch 0 0.00 0.00 0 0 0 0
Roughly the same number of rows, but a big difference in cpu and
elapsed. XP is much faster!!!
When I run the workload simulation setup from Kyte's book 'Effective
Oracle by design' (p. 314) the linux box beats the pants of the XP.
But loading the table is a lot slower. Makes me wonder if something in
my Linux setup is bad...
Any ideas?
jan Received on Sun Sep 21 2003 - 15:36:52 CDT