Personally, I've never understood the whole idea behind extended RAC. What's the point? Seems to me, you're taking an already complex technology (RAC), and adding another layer of complexity (the extended distance between nodes). One of the more performance sensitive areas of a RAC configuration can be latency of the interconnect. Extended distance between nodes will only increase the latency. (You can't argue with the laws of physics.) What's the distance between sites? Also, what's the storage configuration going to look like? You going to have storage replicated at both sites? And, finally, what's the (perceived) benefit? Why do an extended RAC cluster? Seems to me, you could do a RAC at each data center, and do DataGuard between the sites. Any reason such a configuration would not be sufficient? I've never worked with an extended RAC configuration, so, I don't have any first hand experience, but, in my view, it seems like a lot of added headache for very little (if any) benefit. Hope that helps, -Mark ________________________________________ From: oracle-l-bounce@xxxxxxxxxxxxx [oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Avadhani mys [avadhanimys@xxxxxxxxx] Sent: Thursday, August 05, 2010 6:05 To: oracle-l; oaugnetdba-on@xxxxxxxx Subject: Extended RAC:11G R2 Hello Gurus, Has anyone implemented extended RAC, please let me know the pro's and con's and any useful links would be really helpful.We are in the process of finalizing proposal to implement extended RAC. Thanks in Advance --Avadhani -- //www.freelists.org/webpage/oracle-l