BladeSystem Virtual Connect
cancel
Showing results for 
Search instead for 
Did you mean: 

Any concerns raised by your Network/Telcomm Group about using VC?

macblade
Occasional Visitor

Any concerns raised by your Network/Telcomm Group about using VC?

We are current blades users, but use integrated fibre and ethernet pass thru. Our Network/Telecomm Group has not supported a move to virtual connect in the past due to a concerns about support and manageability. While we feel these are not well founded, we wanted to see if others in the community have experienced similar concerns and what they did to overcome them. Thanks very much...
2 REPLIES
stevevr
Occasional Visitor

Any concerns raised by your Network/Telcomm Group about using VC?

Hi there, The question:"Is it a device they we have to manage or not?", is one of responsability/accountability. Saying yes, you loose a part of your departments flexibility and a piece of your enclosure becomes a sort of blackbox to you, changes to the VC needs to be asked using the proper channels etcetc... This is, imho a downside but can be turned into an upside also. You don't have to support it/upgrade it. Keep in mind however that in a blade enclosure everything is tied together. This means you update the VC ethernet and it also impacts the VC fibre channel modules in some cases. So a danger lurks behind the corner. You could for example let them take responsability for the config and support of the device, BUT keep control in saying that every change to the config needs to be initiated by you. That way they are covered :) However I think it's important to keep the control of the VC in your department. You don't need it to manage your enclosure but it helps ALOT. A good setup needs to be tested and validated with them, once that's done, nothing stands in your way of using this tech. Well just my 2c...
ilouser
Occasional Advisor

Any concerns raised by your Network/Telcomm Group about using VC?

We're in the throes of rolling out a C7000 setup with VC 1-10GB and VC Flex-10 modules. Our approach is to have the entire config locked down on a unix machine and it can take the machine from bare metal to fully configured in a few minutes. That way it's easy to fix things and run verifications to see if anything has changed. In this way when we hand it over to Ops to run we can give them simple CLI commands to maintain the machines, keeping them away from the internals as much as possible. We love them dearly but no way will we want them to be driving a web browser to add in enet-connections or the like.