Why would this be better than having the same five nodes all sharing the workload?
I can say why it’s worse: in the event of a failure, there’s more data to copy which means the system would take longer to rebalance and therefore the window where you have no redundancy is larger. The closest to the setup you describe would be firing up a new node when one fails (as happens more-or-less automatically if you use something like Kubernetes). > On 6 Dec 2019, at 11:31, ashishb888 <[email protected]> wrote: > > We have 4 nodes (with attribute data-node) which are used for data storage. > What we want is to have an additional node (with attribute data-node) and do > not use it until one of the node (among the 4 nodes) goes down. > > > > -- > Sent from: http://apache-ignite-users.70518.x6.nabble.com/
