This is a question for someone with a lot of experience networking application servers or for a network engineer who worked intimately to support a team or someone who did.
If I were going to roll out a new suite of applications (JD Edwards EnterpriseOne , for example) , what considerations would I have to make with regard to network architecture and connectivity?
For example, creating a dual-homed server architecture for NIC/switch-port redundancy. Or creating a management vlan, separate from the data vlan, and using a separate NIC for that. Or perhaps placing an application server that has to constantly sieze information from an SQL database server on the same vlan/subnet to avoid layer 3 switching (routing) between hosts on different vlans. Or maybe placing the production servers and dev servers in separate vlans, etc.
The reason I am asking is that I am going to have to help an applications support group make network architecture decisions when they roll out the new suite of applications and associated servers.
From a networking perspective, all I have to do is configure an L2 server farm access switch and trunk the vlans up to a distribution layer. But I want to help them make the decision regarding how many vlans to create and how to subnet them, etc.