I am reorienting myself to Hyper-V in its Windows Server 2012 R2 incarnation and have spent a fair bit of time reading Microsoft's Designing Your Cloud Infrastructure documentation. I am in the SME space and working with a pretty typical deployment: three to four Dell R710s as Hyper-V hosts, an EqualLogic PS6100 for shared storage, and 1Gb for traditional data and SAN traffic with physically discrete switches for each network.
The existing infrastructure is a VMware vSphere implementation but I think it is close to what Microsoft is calling a Non-Converged Data Center Configuration: 1) Separate networks for storage, vMotion/Live Migration, Management and Guest traffic and 2) traditional SAN storage. This sounds like the appropriate approach but I do not want to rule out a design architecture simply because it is new and I do not fully understand it.
My only other experience with converged networking is with Cisco UCS and vSphere so to me "converged networking" means the entire networking infrastructure is basically stateless and virtualized. Microsoft's converged networking seems to mean "using Storage Spaces".
My goals for this deployment are:
- Simple - I don't want to overbuild or make it unnecessarily complicated. Having the appropriate resources is hard in the SME space.
- Flexible. I want there to be a lot of room for growth and scale in the future. The company is growing, I want the infrastructure to be ready to grow with it.
- Clean. It can be simple and small and slow but it needs to be clean. No hackery allowed. Small environments can be professional too.
- Ease of transition - I want to be able to transition from the vSphere implementation live without doing a cutover. I would like to have both environments up at the same time.
Why would I want to use a converged networking design when I have traditional SAN storage and am limited to 1Gb? What advantages and disadvantages does it have over non-converged virtualization networks for the SME space?