IBM Websphere Question

IamtheWalrus wrote: »

I am new to Websphere and would like if someone could help clarify my understanding of my set-up in my office.

I have a web application that has 2 clusters, with 2 nodes in each cluster.

I know the set-up is deployed on 2 machines; A and B. I know that both machines are involved in each cluster, e.g.

Cluster 1 has a node from A and one from B.
Cluster 2 has a node from A and one from B also.

I'm a little confused at how the set-up works. In the above, how many Websphere application servers are there? What is a cell?

Any help is very welcome.

Nodes, etc. in WAS are a bit of an abstract concept, but ultimately the hierarchy of organisation goes cell (top level) -> node -> server. In parallel with this, a cluster is a grouping of application servers, each of which is configured on/within a node, resulting in the alternative hierarchy of cell -> cluster -> server. So a cluster is a logical grouping of application servers, while a node is more of a physical location for these servers. The cell is the top level organisation "container" or scope and everything exists within that, albeit most stuff is defined at sub-scopes.

In your specific example, it sounds as though you have one cell, with 2 nodes and 2 clusters. Each cluster has a server physically located within each of the 2 nodes, i.e. you have the servers cluster1node1, cluster1node2, cluster2node1 and cluster2node2. To be really precise about it, clusters do not contain nodes, but they contain servers which exist within a node, if that makes sense?