Hi, I'm trying to write a simple benchmark to emulate oracle redo logs. What I want to do is to measure the raw theoretical performance that I could achieve from a disk configuration when they are used for redo logs.
If I have understood it correctly Oracle running on Solaris writes a sequential stream of 512 byte blocks to disk. Is it really that simple? So this could be measured with a simple dd command like this $ dd if=/dev/zero of=redo.1 bs=512 count=2097152 It seems to correlate with what I'm seeing on our production systems at the moment but I don't have enough Oracle knowledge to be sure. cheers, Nickus -- Have a look at my blog for sysadmins! http://aspiringsysadmin.com _______________________________________________ perf-discuss mailing list perf-discuss@opensolaris.org
