The other thing is that you created your datafiles at 32k- I am forgetting how orion works but if your OLTP is using a different block size (8k ? ) this might cause some anomaly: The vast majority of my projects use 8k block sizes. Just make sure to test orion using the same block size as your planned db to maintain results that are comparable. -----Original Message----- From: Austin Hackett [mailto:hacketta_57@xxxxxxx] Sent: Wednesday, June 20, 2012 2:44 PM To: CRISLER, JON A Cc: Oracle-L@xxxxxxxxxxxxx Subject: Re: Orion, NFS, & File Size Hi Jon Thanks for your help on this... Attempt 1 orion.lun was: /orion_test/test1.txt Attempt 2 orion.lun was: /orion_test/test1.txt /orion_test/test2.txt /orion_test/test3.txt /orion_test/test4.txt /orion_test/test5.txt Attempt 3 orion.lun was: /orion_test/test1.txt /orion_test/test2.txt /orion_test/test3.txt /orion_test/test4.txt /orion_test/test5.txt /orion_test/test6.txt /orion_test/test7.txt /orion_test/test8.txt /orion_test/test9.txt /orion_test/test10.txt Attempt 4 was the same as 3 but with -num_disks 24 passed at the command line. /orion_test was mounted with the Oracle/NetApp recommended mount options for RAC data files I just remembered reading this thread when doing my research recently: //www.freelists.org/post/oracle-l/ORION-num-disks,12 It seems to chime with what I'm seeing, so could be another approach to try... On 20 Jun 2012, at 19:16, CRISLER, JON A wrote: > What is the content of your lun file? These results seem in line with > 1 thread on a 1 gig enet link- do you have all of the datafiles in > your config file? -- //www.freelists.org/webpage/oracle-l