Yong,I suppose that there may be some difference between 4 and 6 nodes with respect to the occurrence of 2 vs 3-way block negotiations. However, I don't think that will be a noticeable. I think the only time you'd have enough impact to really measure the difference would be between 2 and 3 nodes and that's supported by the study I mentioned (that I still can't seem to locate).
Dan Yong Huang wrote:
Dan's theory says in case of 6 nodes, there's relatively more 3-way traffic, which could slow down the overall performance. Correct, Dan?From: Dan Norris <dannorris@xxxxxxxxxxxxx> ... In any case where a block is cached in another instance, there are three parties involved: the requestor, the holder, and the global resource master for the requested block. In a two-node cluster, two of these three parties are the same instance and therefore the messaging between them is very fast. When a third node is added, there's a statistical probability that these three parties will involve all three instances for a significant portion of the occurrencesof this situation. In the case where three instances are involved,the communications take longer since all messaging has to cross theinterconnect.
-- //www.freelists.org/webpage/oracle-l