║JAI SRI GURUDEV║

Sri AdichunchanagiriShikshana Trust (R)

SJB INSTITUTE OF TECHNOLOGY

BGS Health & Education City, Kengeri , Bangalore – 60 .

DEPARTMENT OF COMPUTER SCIENCE & ENGINEERING

Cloud Computing (21CS72)

by

Dr. Krishna A N
Professor & Head
Department of Computer Science &
Engineering
SJB Institute of Technology, Bangalore
 Child partitions are used to host guest operating systems
and do not have access to the underlying hardware, but
their interaction with it is controlled by either the parent
partition or the hypervisor itself.

Microsoft Hyper-V: Architecture

 Hyper-V is an infrastructure virtualization solution developed by Microsoft
for server virtualization. It uses a hypervisor-based approach to hardware
virtualization, which leverages several techniques to support a variety of
guest operating systems.
 Hyper-V supports multiple and concurrent execution of guest operating
systems by means of partitions. A partition is a completely isolated
environment in which an operating system is installed and run.
 Hyper-V takes control of the hardware, and the host operating system
becomes a virtual machine instance with special privileges, called the
parent partition.
 The parent partition (also called the root partition) is the only one that has
direct access to the hardware. It runs the virtualization stack, hosts all the
drivers required to configure guest operating systems, and creates child
2 partitions through the hypervisor.
 Architecture

3
 Architecture: Hypervisor
 The hypervisor is the component that directly manages the underlying
hardware (processors and memory). It is logically defined by the following
components:
 Hypercalls interface-This is the entry point for all the partitions for the execution of
sensitive instructions. This is an implementation of the paravirtualization approach.
This interface is used by drivers in the partitioned operating system to contact the
hypervisor using the standard Windows calling convention. The parent partition also
uses this interface to create child partitions.
 Memory service routines (MSRs)- These are the set of functionalities that control
the memory and its access from partitions. The hypervisor uses the Input/Output
Memory Management Unit (I/O MMU or IOMMU) to fast-track access to devices
from partitions by translating virtual memory addresses.
 Advanced programmable interrupt controller (APIC)- This component represents the
interrupt controller, which manages the signals coming from the underlying
4 hardware when some event occurs (timer expired, I/O ready, exceptions and traps).
 Architecture: Hypervisor
 Each virtual processor is equipped with a synthetic interrupt controller (SynIC), which
constitutes an extension of the local APIC. The hypervisor is responsible of
dispatching, when appropriate, the physical interrupts to the synthetic interrupt
controllers.
 Scheduler- This component schedules the virtual processors to run on available
physical processors. The scheduling is controlled by policies that are set by the
parent partition.
 Address manager. This component is used to manage the virtual network addresses
that are allocated to each guest operating system.
 Partition manager. This component is in charge of performing partition creation,
finalization, destruction, enumeration, and configurations. Its services are available
through the hypercalls interface API previously discussed.
 The hypervisor runs in Ring -1. By executing in this highly privileged mode,
the hypervisor can support legacy operating systems that have been
5 designed for x86 hardware.
 Architecture: Enlightened I/O and synthetic
devices
 Enlightened I/O provides an optimized way to perform I/O operation.
Enlightened I/O leverages VMBus, an interpartition communication channel
that is used to exchange data between partitions (child and parent) and is
utilized mostly for the implementation of virtual device drivers for guest
operating systems.
 There are three fundamental components: VMBus, Virtual Service Providers
(VSPs), and Virtual Service Clients (VSCs).
 VMBus implements the channel and defines the protocol for
communication between partitions.
 VSPs are kernel-level drivers that are deployed in the parent partition and
provide access to the corresponding hardware devices. These interact with
VSCs, which represent the virtual device drivers (also called synthetic
drivers) seen by the guest operating systems in the child partitions.
6
 Architecture: Enlightened I/O and synthetic
devices
 Operating systems supported by Hyper-V utilize this preferred
communication channel to perform I/O for storage, networking, graphics,
and input subsystems. This also results in enhanced performance in child-
to-child I/O as a result of virtual networks between guest operating
systems.

7
 Parent partition
 The parent partition executes the host operating system and implements the
virtualization stack that complements the activity of the hypervisor in
running guest operating systems.
 This partition always hosts an instance of the Windows Server 2008 R2, which
manages the virtualization stack made available to the child partitions.
 This partition is the only one that directly accesses device drivers and
mediates the access to them by child partitions by hosting the VSPs.
 The parent partition is also the one that manages the creation, execution,
and destruction of child partitions by means of the Virtualization
Infrastructure Driver (VID), which controls access to the hypervisor and allows
the management of virtual processors and memory.
 For each child partition created, a Virtual Machine Worker Process (VMWP) is
instantiated in the parent partition, which manages the child partitions by
8 interacting with the hypervisor through the VID.
 Child partitions
 Child partitions are used to execute guest operating systems. These are
isolated environments that allow secure and controlled execution of guests.
 Two types of child partition exist, they differ on whether the guest
operating system is supported by Hyper-V or not.
 These are called Enlightened and Unenlightened partitions, respectively.
The first ones can benefit from Enlightened I/O; the other ones are
executed by leveraging hardware emulation from the hypervisor.

9
 Cloud computing and infrastructure
management
 Hyper-V constitutes the basic building block of Microsoft virtualization
infrastructure.
 To increase the performance of virtualized environments, a new version of
Windows Server 2008, called Windows Server Core, has been released. This is
a specific version of the operating system with a reduced set of features and a
smaller footprint.
 In particular, Windows Server Core has been designed by removing those
features, which are not required in a server environment, such as the GUI
component and other bulky components such as the .NET Framework and all
the applications developed on top of it (for example, PowerShell).
 Adv.: It allows for reduced maintenance (i.e., fewer software patches),
reduced attack surface, reduced management, and less disk space.
 Disadv.: Reduced features. Still, there is the opportunity to leverage all the
10 “removed features” by means of remote management from a fully featured
 Cloud computing and infrastructure
management
 Another component that provides advanced management of virtual
machines is System Center Virtual Machine Manager (SCVMM) 2008. This is
a component of the Microsoft System Center suite, which brings into the
suite the virtual infrastructure management capabilities from an IT lifecycle
point of view.
 Essentially, SCVMM complements the basic features offered by Hyper-V
with management capabilities, including:
– Management portal for the creation and management of virtual instances
– Virtual to Virtual (V2V) and Physical to Virtual (P2V) conversions
– Delegated administration
– Library functionality and deep PowerShell integration
– Intelligent placement of virtual machines in the managed environment
– Host capacity management
11