Introduction to Compute

Virtualization – Part 2
Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Foreword

⚫ Cloud computing 1.0 focused on virtualization which today has become

the foundation of cloud computing. This chapter focuses on the
architecture and technologies involved in compute virtualization.

Page 1 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Objectives

⚫ Upon completion of this chapter, you will:

 Be able to describe what I\O virtualization is.

 Understand the differences between virtualization and cloud computing.

 Understand KVM technology.

Page 2 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Contents

1. Introduction to Virtualization

 CPU Virtualization

 Memory Virtualization

 I/O Virtualization

2. Understand the differences between virtualization and cloud

computing.

3. Understand KVM technology.

Page 3 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ With compute virtualization, a large number of VMs can be created on a single

host, and these VMs all need to access the I/O devices of this host.

⚫ However, I/O devices are limited. I/O device sharing among multiple VMs requires
VMM. VMM intercepts access requests from VMs to I/O devices, simulates I/O
devices using software, and responds to I/O requests. This way, multiple VMs can
access I/O resources concurrently.

Page 4 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ I/O virtualization can be implemented in the following methods:

 Full virtualization,

 Paravirtualization,

 Hardware-assisted virtualization

Page 5 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ Emulation [Full Virtualization]:

⚫ VMM virtualizes I/O devices for VMs. When a VM initiates an I/O request to an I/O
device, VMM intercepts the request sent by the VM, and then sends the real
access request to the physical device for processing.

⚫ No matter which type of OS is used by the VM, the OS does not need to be
modified for I/O virtualization.

Page 6 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ Emulation [Full Virtualization]:

⚫ VMM needs to intercept I/O requests delivered by each VM in real time and
emulates the request to a real I/O device. Real-time monitoring and emulation are
implemented by software programs on the CPU, which causes severe
performance loss to the server.

⚫ Complete simulation of the hardware, for example, keyboard and mouse. Access
to such hardware depends on the capture of focus by the host, and leads to poor
performance in some cases.

Page 7 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ Paravirtualization:

⚫ Paravirtualization requires each VM to run a frontend driver. When VMs need to

access an I/O device, the VMs send I/O requests to the privileged VM (act as
hypervisor) through the frontend driver, and the backend driver of the privileged
VM collects the I/O request sent by each VM.

⚫ Access to hardware drivers is transferred from the I/O frontend to the I/O backend.
This mode is usually only used for hard disks and Network Interface Cards
(NICs) and delivers high performance.

Page 8 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ Front-Ends and Back-Ends

⚫ Front-End:

 Guest driver and emulated device

⚫ Back-End:

 Device implementation

 Host physical resources

Page 9 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ Paravirtualization:

⚫ This reduces the performance loss of VMM and therefore delivers better I/O
performance.. However, the VM OS needs to be modified (usually Linux).
Specifically, the I/O request processing method of the OS needs to be changed
so that all the I/O requests can be sent to the privileged VM for processing.

⚫ Guest uses specialized driver for optimized virtual hardware. Drivers must be
implemented for each type of OS.

Page 10 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ IO-through (Hardware-assisted virtualization):

⚫ Hardware-assisted virtualization directly installs the I/O device driver in the VM OS

without any change to the OS. This method is equivalent to traditional PC OS
access to hardware.

⚫ Therefore, the time required for a VM to access the I/O hardware is the same as
that for a traditional PC to access the I/O hardware.

Page 11 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ IO-through (Hardware-assisted virtualization):

⚫ Hardware-assisted virtualization is like an intelligent information collection and

processing platform. Users' requests can be directly submitted to the platform and
the platform automatically processes the requests.

⚫ Therefore, hardware-assisted virtualization outperforms full virtualization and

paravirtualization in terms of I/O performance. However, hardware-assisted
virtualization requires special hardware support.

Page 12 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization

⚫ IO-through (Hardware-assisted virtualization):

 Guest can access the device directly

 Virtually eliminates all emulation and back-end overhead

 Each device is limited to use by one VM

 Introduces strong coupling between Guest and hardware

Page 13 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 I/O Virtualization
⚫ I/O Virtualization:

⚫ Emulation (Full Virtualization)

 Best option for correctness and abstraction

 High performance cost

⚫ Paravirtualization

 Optimize driver and virtual device interaction

 Guest is “aware” of virtualization

⚫ IO-through

 Best option for performance

 Strong coupling with hardware

Page 14 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Cloud Computing and Virtualization

Cloud
Computing

Page 15 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Cloud Computing and Virtualization

⚫ CPU virtualization, memory virtualization, and I/O virtualization can be

implemented to enable the reuse of physical resources. Multiple virtual servers can
run on a physical host at the same time, and each virtual server can run different
workloads. This improves hardware utilization.

⚫ In cloud computing, we mainly use virtualization to implement IaaS cloud

services. There are three cloud service models: IaaS, PaaS, and SaaS. Some PaaS
and SaaS services are implemented based on virtualization

Page 16 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Cloud Computing and Virtualization

⚫ Cloud computing is a business model that provides users with IT services anytime
anywhere. Virtualization is an important technical means for cloud computing
implementation.

⚫ There are many mainstream virtualization technologies. Generally, open-source

and closed-source are used for classification. Open-source technologies include
KVM and Xen. Closed-source virtualization technologies include Microsoft Hyper-
V, VMware vSphere, and Huawei FusionSphere.

Page 17 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Cloud Computing and Virtualization

⚫ Open-source technologies are free of charge and can be used anytime. Users can
customize some special requirements based on open-source code. Once a
problem occurs in the system, the system recovery strongly relies on the
administrator's skillset and experience.

⚫ In closed-source technologies, users cannot view or customize source code.

Closed-source virtualization products are generally not free of charge and can be
used out of the box. If a system problem occurs, vendors provide all-round
support.

Page 18 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Mainstream Compute Virtualization
Technologies
Compute Virtualization
CPU virtualization, memory virtualization,
I/O virtualization

Open Source Closed Source

KVM Hyper-V

Xen VMware ESXi

Huawei
FusionSphere

Page 19 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 What’s KVM?

⚫ Kernel-based Virtual Machine (KVM) is a Type-II full virtualization solution. It is a

Linux kernel module. A physical machine with a Linux kernel module installed can
function as a hypervisor, which does not affect the other applications running on
the Linux OS. Each VM is one or more processes. You can run the kill command to
kill the processes.

Page 20 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 What’s KVM?

Other Guest OS Guest OS

Application

QEMU QEMU

KVM
Linux
HW Drivers

Hardware

Page 21 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 What’s KVM?

⚫ After the KVM module is installed in a common Linux OS, three running modes are
added:

 Guest Mode: VMs, including their CPUs, memory, and disks, run in a restricted
CPU mode.

 User Mode: The quick emulator (QEMU) typically runs in this mode. QEMU
emulates I/O requests.

 Kernel Mode: In this mode, the hardware can be operated. When the guest OS
executes an I/O operation or privileged instruction, a request needs to be
submitted to the user mode, and then the user mode initiates a hardware
operation request to the kernel mode again to operate the hardware.

Page 22 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 What’s KVM?

VM VM
Application program A Application program B

Binaries/Libraries Binaries/Libraries
User
Guest OS Guest OS
Space User
ProcessesSpace QEMU QEMU
Processes

User space
Kernel space
KVM
Linux kernel
module

Physical
hardware

Page 23 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM Architecture

⚫ QEMU was not a part of KVM. It was a universal open-source virtualization

emulator that uses pure software to implement virtualization. The guest OS
considers that it is interacting with hardware.

⚫ Actually, QEMU is interacting with hardware. This means that all interactions with
the hardware need to pass through QEMU. Therefore, the simulation
performance delivered by QEMU is low. QMEU is able to simulate CPUs and
memory. In KVM, only QEMU is used to simulate I/O devices.

Page 24 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM Architecture

⚫ The KVM kernel module is the core of a KVM VM. This module initializes the CPU
hardware, enables the virtualization mode, runs the guest machine in the VM mode,
and supports the running of the virtual client.

⚫ However, a VM requires other I/O devices such as Network Interface Cards (NICs)
and hard disks besides CPUs and memory. QEMU is required to implement other
virtualization functions.

⚫ Therefore, the KVM kernel module and QEMU form a complete virtualization
technology.

Page 25 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM I/O Process - Default

Guest OS
4
I/O Shared Page 1 10
QEMU I/O 7 Device
Emulation Code
Driver

3 8
Notification 2 9

I/O Trap
5 Code
KVM Module
Linux Kernel
Device
Driver

Physical Hardware

Page 26 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM I/O Process - Default

⚫ Let's see how VM disk operations are performed:

1. A disk device of a VM initiates an I/O operation request.

2. I/O Trap Code (I/O capture program) in the KVM module captures the I/O
operation request, performs corresponding processing, and then puts the processed
request into the I/O shared page.

3. The KVM module notifies QEMU that a new I/O operation request is placed in the
shared page.

4. After receiving the notification, QEMU obtains the detailed information about the
I/O operation request from the shared page.

Page 27 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM I/O Process - Default

5. QEMU simulates the request and calls the device driver running in kernel mode
based on the request information to perform the real I/O operation.

6. The I/O operation is then performed on physical hardware through the device
driver.

7. QEMU returns the operation result to the shared page and notifies the KVM
module that the I/O operation is complete.

8. I/O Trap Code reads the returned result from the shared page.

9. I/O Trap Code returns the operation result to the VM.

10. The VM returns the result to the application that initiated the operation.

Page 28 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM I/O Process - Virtio

⚫ In steps 2, 3, and 7 (Default) , KVM does not make any modification on the I/O
operation except for capturing the request and sending the notification.

⚫ The Virtio technology was developed to simply this procedure.

Page 29 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM I/O Process - Virtio

Guest OS
QEMU
1 8
3 2
Virtio Backend Virtio Frontend
Driver vring Driver
6 7

Notification
I/O Trap
4 Code
KVM Module
Linux Kernel
Device Driver

Physical Hardware

Page 30 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM I/O Process - Virtio

⚫ If Virtio is used, the procedure is as follows:

1. The VM initiates an I/O operation request.

2. The I/O operation request is not captured by the I/O capture program.
Instead, the request is stored in the ring buffer between the frontend and
backend drivers. At the same time, the KVM module notifies the backend
driver.

3. QEMU obtains the detailed information about the operation request from
the ring buffer.

Page 31 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM I/O Process - Virtio

4. The backend driver directly calls the actual physical device driver to perform the
I/O operation.

5. The operation is completed by the device driver.

6. QEMU returns the operation result to the ring buffer, and the KVM module notifies
the frontend driver.

7. The frontend driver obtains the operation result from the ring buffer.

8. The frontend driver returns the result to the application that initiated the
operation.

Page 32 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM I/O Process - Virtio

⚫ The advantages of Virtio are as follows:

 Saves the hardware resources required for QEMU emulation.

 Reduces the number of I/O request paths and improves the performance of
virtualization devices.

⚫ Virtio has some disadvantages. For example, some old or uncommon devices
cannot use Virtio but can only use QEMU.

Page 33 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM Architecture

Management Tools

Virt-
Virsh Virt-viewer Virt-install Others
manager

Libvirt
A KVM system consists of three parts: KVM
kernel module, QEMU, and management
tool.
Linux Kernel
(KVM module)

QEMU QEMU

Guest Guest
OS OS

Page 34 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM Architecture
Libvirt

VM VM VM VM
(Guest OS) (Guest OS) (Guest OS) (Guest OS)
Manage
-ment
Hypervisor Hypervisor Tools Hypervisor
Libvirt
Linux host Linux host

Node Node

Libvirt is an open-source project and is a powerful management

tool. It is able to manage virtualization platforms such as KVM.

Page 35 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 KVM Architecture
Libvirt
⚫ In cloud computing, there are various hypervisors. Each hypervisor has its own
parameters that are complex and difficult to use. Hypervisors are not unified, and
there is no unified programming interface to manage them, which severely affects
the cloud computing environment.

⚫ With Libvirt, it can connect to various hypervisors, such as KVM and Xen, and
provide APIs in various languages. Libvirt serves as the middle layer between the
management tool and hypervisor and is completely transparent to upper-layer
users.

Page 36 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 The Xen Hypervisor Approach

Page 37 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 The Xen Hypervisor Approach

⚫ Xen is an open source hypervisor program developed by Cambridge University. It

creates logical pools of system resources so that many virtual machines can share
the same physical resources.

⚫ Xen is a type 1 hypervisor that runs directly on the system hardware. Xen inserts a
virtualization layer between the system hardware and the virtual machines, turning
the system hardware into a pool of logical computing resources that Xen can
dynamically allocate to any guest operating system.

⚫ The operating systems running in virtual machines (Guest OS) interact with the
virtual resources as if they were physical resources.

Page 38 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 The Xen Hypervisor Approach

⚫ Xen is based on paravirtualization. ; it requires that the guest operating systems

be modified to support the Xen operating environment. However, the user space
applications and libraries do not require modification.

⚫ This is probably done by removing the performance loss while executing the
instructions requiring significant handling and by modifying portion of the guest
operating system executed by Xen, with reference to the execution of such
instructions.

Page 39 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 The Xen Hypervisor Approach

⚫ In paravirtualization (PV), the guest OS knows that it is running in a virtual

machine and cooperates with the hypervisor to get access to the actual hardware.

⚫ The guest OS asks the hypervisor to perform functions that would normally
require direct access to hardware: e.g., accessing certain CPU registers, etc. It is
much easier for the hypervisor to translate these calls instead in case of full
virtualization that emulates hardware devices and network interfaces to guests.

⚫ The result of Paravirtualization (PV) is a very lightweight and fast hypervisor.

Page 40 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 The Xen Hypervisor Approach

⚫ Xen does not provide any device drivers. It has no direct knowledge of networking,
external storage devices, video, or any other common I/O functions found on a
computing system. But provides a mechanism by which a guest operating system
can be given direct access to a physical device.

⚫ Operating system modifications are necessary for reasons like:

 So that Xen can replace the operating system as the most privileged software.

 Xen contains a set of paravirtualized (PV) drivers that the guest loads instead of
the actual hardware drivers. So that Xen can use more efficient interfaces (such
as virtual network interfaces) to emulate devices — this increases performance.

Page 41 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Components of the Xen hypervisor

⚫ Xen refers to each virtual machine that runs on a system as “domains”.

⚫ When Xen is first installed, it automatically creates the first domain, a special
domain known as Domain 0 (or dom0), is responsible for controlling the
hypervisor and starting other guest operating systems. It is a privileged domain
that can access the hardware resources and can manage all the other domains
(e.g., create, destroy, save, restore, etc.)

⚫ Other guest operating systems are called DomUs domain, where the U stands
for unprivileged. This is because they cannot control the hypervisor or start/stop
other domains.

Page 42 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 The Xen Architecture

Dom0 DomU DomU DomU

PV
Backends

HW PV PV PV
Drivers Frontends Frontends Frontends

Xen

Hardware

Page 43 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Xen Terminologies

• Domain: The term used by Xen to refer to a virtual machine instance.

• Dom0: The primary domain in a Xen system.

• DomU: All the other domains in a Xen system.

• Host: The system that provides the environment for running virtual machines.

• Guest: The virtual machine instance that runs on a host system.

• PVM: Paravirtualized Virtual machine where the Xen domains can only run
modified operating systems.

• HVM: Hardware Virtual Machine where the Xen domains can run unmodified
operating systems such as Microsoft Windows leveraging special virtualization
hardware
Page 44 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Components of the Xen hypervisor

⚫ Domain 0 (Dom0): is the management domain and is responsible for managing

the system. It runs a guest operating system and is responsible for the hardware
devices.

 Dom0 is a privileged domain with direct access to the hardware. It has exclusive
privileges to access the Xen hypervisor and manage all aspects of Guests.

 Dom0 handles all access to the hardware and I/O and manages them on behalf
of the users’ VMs. This is because the hypervisor doesn’t contain device
drivers. Dom0 then shares these resources with guest operating systems.

 The OS running on dom0 has the device drivers and performs I/O operations
on behalf of unprivileged guest domains (domU);

Page 45 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Components of the Xen hypervisor

⚫ The Domain U Guest is a modified open-sourced OS (ex: Linux, Solaris, FreeBSD

or other UNIX system) that is aware of virtualization (no direct access to hardware).

⚫ No rights to directly access hardware resources, unless especially granted.

Page 46 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Components of the Xen hypervisor

⚫ To implement paravirtualization, each paravirtualized datapath consists of two

parts:

1) Backend: that lives in dom0, which provides the virtual device.

2) Frontend: driver within the guest domain, which allows the guest OS to access
the virtual device.

⚫ The backend and frontend use a high-speed software interface based on shared
memory to transfer data between the guest and dom0.

Page 47 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 The Xen approach
• Two drivers are included in Domain 0 to attend requests from Domain UPV or HVM guests
to Access physical I/O devices.

• The Domain U HVM Guest is a native OS with no notion of virtualization (sharing CPU time
and other VMs running). An unmodified OS doesn’t support the Xen split device driver,
Xen emulates devices by borrowing code from QEMU
Domain 0
PV Communicates directly with the local
networking hardware to process all virtual
Network backend
machines requests
driver
Communicates with the local storage disk to
Block backend driver
read and write data from the drive based
upon Domain U requests
HVM
Supports HVM Guests for networking and
Qemu-DM
disk access requests

Page 48 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 The Xen approach

• Domain UPV has access to hardware through front-end drivers using the split device
driver model.

• Shared memory is used for the communication between a domU and dom0

Domain U - PV
Similar to a registry
Console driver

XenStore driver Communicates with the Network backend

driver in Domain 0
Network front-end driver

Block front-end driver Communicates with the Block backend

driver in Domain 0

Page 49 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Xen vs. KVM
⚫ In open-source virtualization technologies, Xen is on a par with KVM. KVM is full
virtualization, while Xen supports both paravirtualization and full virtualization. KVM, a
module in the Linux kernel, is used to virtualize CPUs and memory. It is a process of the
Linux OS.

⚫ Different from KVM, Xen directly runs on hardware, and VMs run on Xen.

⚫ VMs in Xen are classified as the privileged VM (Domain 0) that has the permission to
directly access hardware and manage other VMs (for example, Domain U). Domain 0 must
be started before other VMs. Domain U is a common VM and cannot directly access
hardware resources.

⚫ All operations on Domain U must be forwarded to Domain 0 through frontend and

backend drivers. Domain 0 completes the operations and returns the results to Domain U.

Page 50 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Xen vs. KVM Architecture

Page 51 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Xen vs. KVM (I\O Process)

Page 52 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
 Summary

⚫ This chapter provides basic knowledge about compute virtualization, including

CPU, memory, and I/O virtualization. It introduces you to open-source
virtualization technology by using KVM and Xen as an example.

Page 53 Copyright © 2019 Huawei Technologies Co., Ltd. All rights reserved.
Thank You
www.huawei.com