Virtualization Solution Types
Virtualization solutions basically falls under three categories, and these categories are related to the architecture that is used for the virtualization of the server. The key differences between these architectures are directly connected to the relationship that exists among the hardware, and the layer for virtualization.
Virtual Machine Monitor (VMM) is a virtualization software layer which offers the ability to generate numerous instances which are isolated but share identical hardware resources. The Type-2 VMM architecture may be exemplified through the JVM, or Java Virtual Machines.
The goal of virtualization is to generate the runtime environment inside, which the process may use for the execution of a collection of instructions without need to be reliant on the host system. In this situation, the isolation occurs for distinct processes, and it allows one application to operate on numerous operating systems without the need to worry about the OS dependencies. Server virtualization is not connected to this. The Type-1 VMM, as well as the Hybrid VMM, are the two methods that are used the most.
The Hybrid VMM is a type of stage in which the VMM operates in conjunction with the host OS, and it helps in the creation of the virtual machines. Some examples of the Hybrid VMM include the Microsoft Virtual Server, as well as the VMware Workstation. You must note that despite the fact these solution types are great for the client scenario in which you are only running the virtual machines a fraction of the time, the VMMs can offer a great deal of overhead, and it is not useful for workloads which are resource-intensive.
Virtualization Server Type Functions
With the Type-1 VMM structure, the VMM layer operates directly above the hardware. Many people refer to this layer as being the hypervisor. When the architecture was first created in the 1960s for the mainframe machines, few realized that it would be decades before it was made available on the x86 platforms. This is now available for a variety of different platforms.
There are many solutions available in which the hypervisor is embedded within the firmware. Despite this, it should be thought of as the packaging option, and it does not bring many changes to the technology which underlies it.
When you take a close look at the Type-1 VMMs, you will find that there are basically two primary ways in which you can structure the solutions, and while the first option is monolithic, the second is microkernalized. For both of these two approaches, the authentic Type-1 VMMs which have the hypervisor installed may typically be installed right on the physical hardware itself.
The monolithic hypervisor method will host the VMM within one layer, and this also includes many of the components that were required, like the kernel, the device drivers, and the I/O stack. This is the basic approach which is utilized by solutions like the VMware ESX, and the standard mainframe systems.
The microkernalized method makes use of a think hypervisor which is specifically designed to carry out the primary tasks which are necessary for ensuring the isolation of partition, and which involves the management of memory.
The layer does not allow for the I/O stack, and it does not allow for the device drivers. This is the simple method which is connected with Hyper-V. Through this architecture, both the hardware specific driver and the virtualization stack can be found within the partition which is specified for the parent partition. One of the most important things to become familiar with is the Hypervisor.
Windows Hypervisor
To ensure that there is a strong distinction among the different operating systems which is done by the creation of the virtual processors, timers, and the interrupt controllers, the OS will use the virtual resources just as they would normally be used for their physical attributes.
For the Windows hypervisor, which is directly connected to Hyper-V, there are a number of tasks that can be performed, and some of these include the creation of logical partitions, the management of both memory and processor scheduling, the provision of mechanisms for the virtualization of input/out via the partitions, and the enforcement of the memory access rules.
It is also useful for the enforcement of the access rules for the memory, as well as the enforcement of policy for the CPU usage. Additionally, it is responsible for exposing the basic interface which is known as the hypercall. Because it makes use of the microkernelized method, the hypervisor for Windows is quite small, and it is smaller than 1MB in size.
The basic footprint will help in getting rid of and enhancing the total security of the system. One of the requirements for properly operating Hyper-V is that if you have the x64 system, this technology allows for access to a much bigger address space, and offers support for a system with additional memory.
AMD-V is a virtualization solution which is hardware assisted, and it provides a very privileged layer inside the ring architecture. This basically helps to keep the environment for execution within the hypervisor, and this also means that it is separate from the remainder of the system. This also means that Hyper-V is allowed to run the guest OS which is unmodified without the need to incur a large amount of emulation in terms of performance.
