When it comes to choosing a manufacturer there’s a few factors to consider:
- Experience with their hardware?
- Do they have suitable servers and/or storage?
- Do they have hardware monitoring management packs?
- Do they have PRO tips management packs?
A big reason we choose Hyper-V for hardware virtualization is System Center. That’s why two of those points focus on System Center. I’ve previously talked about server hardware monitoring for IBM, Fujitsu, HP and Dell. I’ve no experience with the Fujitsu stuff so I cannot rate it. But that doesn’t stop me from ranking IBM as last in this category, based on my experience. Dell and HP both do a great job in this category. But that’s just the servers. You should make sure your storage and networking solutions also have OpsMgr health and performance management packs.
One of the coolest aspects of Hyper-V/System Center is Performance and Resource Optimization (PRO). PRO management packs can detect failures or performance issues in OpsMgr managed resources in your Hyper-V infrastructure. That alert is passed on to Virtual Machine Manager. VMM then takes an action, i.e. Live Migrate a VM, based on that failure. For example, a NIC might fail in a host server. Failover Clustering will not react to this but PRO will. VM’s might be offline because of this so the VM’s are moved to the most suitable host for each workload. This all means using a Hyper-V cluster.
You can find a listing of PRO management packs on the Microsoft site. It’s an interesting list, including Dell, HP and IBM. Storage components such as QLogic and Emulex (HBA’s) are there. NetApp have a PRO pack. Veeam even have one for VMM managed VMware environments – it seems like you might be able to bring PRO to vSphere!
So here’s what I’d do. Start off by making sure there is an OpsMgr management pack for the hardware. This really should be a free download from the manufacturer. Then make sure the components you are going to be relying on for Hyper-V performance have PRO packs. That’ll allow you to build a management system that builds in intelligence and automation into your virtualization and hardware. You will really be able to change how you do IT; using accurate information to always make the best decisions, even when you’re asleep in bed!