Microsoft Partner Enablement – Hyper-V
November 4, 2013 Leave a comment
There are 3 possible version types to deploy Hyper-V:
- Server with GUI
- Server Core (+ opportunity to install other roles)
- Free standalone Hyper-V Server 2012 (Server core without other roles)
As a hypervisor all three of the options allow the same features and capabilities.
Traditional per server installation options include;
- Deploy from DVD/ISO
- Deploy from USB Stick
- Network Deployment
Traditional Network deployment methods;
- System Center Configuration Manager 2012 SP1 (SCCM)
- Microsoft Deployment Toolkit 2012 U1 (MDT)
- Windows Deployment Services (WDS)
With the traditional network deployments you have various benefits depending on the level of deployment you require, as an example;
- The MDT allows a LiteTouch deployment which effectively means MANY of the options required to deploy and interact with (wizards) are automated during the deployment process.
- The SCCM allows a ZeroTouch deployment which effectively means ALL of the options required to deploy and interact with (wizards) are automated during the deployment process.
The work behind the scenes varies massively between a LiteTouch and ZeroTouch deployment and will need to be considered when implementing, for example, a zero touch deployment would be a great solution in any environment, but this is really looking towards a massive deployment scenario, such as 50+ hosts, whereas a 10 user business may just benefit from a LiteTouch (for ease of redeployment) or physical deployment via media or USB.
The preferred Hyper-V deployment option would be to utilise Virtual Machine Manager – this utilises the SCCM and WDS capabilities but works around a Virtual Infrastructure.
Whichever of the network deployment options you decide on, be mindful that Microsoft Hyper-V is a stateful PXE deployment, at this moment in time VMware Auto Deploy is the only Hypervisor of the two that allows a Stateless option (introduced in vSphere 5.1).
At this point the course went into a comparison between Hyper-V and ESXi and the maximums for scalability, however I’ll summerise as we rarely reach these in any environment and we can reference them if needed.
- Max 320 logical processors and 4TB physical memory per host
- 1024 VM’s per host
- 64 Physical Nodes
- 8000 VM’s
- 64 vProcessors and 1TB RAM
- In-guest **NUMA supported (aligns with host resources to increase performance)
**NUMA – Non-Uniform Memory Access – As defined by Wikipedia –
Non-Uniform Memory Access is a computer memory design used in multiprocessing, where the memory access time depends on the memory location relative to a processor. Under NUMA, a processor can access its own local memory faster than non-local memory (memory local to another processor or memory shared between processors).
Support for Guest OS’s
Windows 8 – 32 Virtual CPUs
Windows 7/SP1 – 4 Virtual CPUs
Vista SP2/XP SP3/XP x64 SP3 – 2 Virtual CPUs
CentOS/Red Hat/SUSE/OpenSUSE/Ubuntu – 64 Virtual CPUs
Thanks for viewing and I hope this post was a helpful reference
Steve – Guru365