On 5/30/07, Richard McDougall <[EMAIL PROTECTED]> wrote:
Unfortunately, dd won't replicate anything close to what Oracle is sensitive to for redo logs...
Thanks for the input Richard. After running my initial tests I quickly realized the same thing. I wrote a simple python script where I can better control the size of the blocks and I do a sync after every write. At least that script gave a number that was pretty close to what I'm seeing in production and I can now run this script on the new array to get a feeling on what to expect. cheers, Nickus _______________________________________________ perf-discuss mailing list perf-discuss@opensolaris.org
