Chapter 5 – Cloud Resource

Virtualization
Contents
 Virtualization.
 Layering and virtualization.
 Virtual machine monitor.
 Virtual machine.
 Performance and security isolation.
 Architectural support for virtualization.
 x86 support for virtualization.
 Full and paravirtualization.
 Xen 1.0 and Xen 2.0.
 Performance comparison of virtual machine monitors.
 The darker side of virtualization.
 Software fault isolation.

Dan C. Marinescu Cloud Computing: Theory and Practice. Chapter 5 2

Motivation
 There are many physical realizations of the fundamental
abstractions necessary to describe the operation of a computing
systems.
 Interpreters.
 Memory.
 Communications links.
 Virtualization is a basic tenet of cloud computing, it simplifies the
management of physical resources for the three abstractions.
 The state of a virtual machine (VM) running under a virtual machine
monitor (VMM) can de saved and migrated to another server to
balance the load.
 Virtualization allows users to operate in environments they are
familiar with, rather than forcing them to idiosyncratic ones.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 3
Motivation (cont’d)

 Cloud resource virtualization is important for:

 System security, as it allows isolation of services running on
the same hardware.
 Performance and reliability, as it allows applications to migrate
from one platform to another.
 The development and management of services offered by a
provider.
 Performance isolation.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 4
Virtualization
 Simulates the interface to a physical object by:
 Multiplexing: creates multiple virtual objects from one instance
of a physical object. Example - a processor is multiplexed
among a number of processes or threads.
 Aggregation: creates one virtual object from multiple physical
objects. Example - a number of physical disks are aggregated
into a RAID disk.
 Emulation: constructs a virtual object from a different type of a
physical object. Example - a physical disk emulates a Random
Access Memory (RAM).
 Multiplexing and emulation. Examples - virtual memory with
paging multiplexes real memory and disk; a virtual address
emulates a real address.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 5
Layering
 Layering – a common approach to manage system complexity.
 Minimizes the interactions among the subsystems of a complex
system.
 Simplifies the description of the subsystems; each subsystem is
abstracted through its interfaces with the other subsystems.
 We are able to design, implement, and modify the individual
subsystems independently.
 Layering in a computer system.
 Hardware.
 Software.
 Operating system.

 Libraries.

 Applications.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 6
Interfaces
 Instruction Set Architecture (ISA) – at the boundary between
hardware and software.

 Application Binary Interface (ABI) – allows the ensemble consisting

of the application and the library modules to access the hardware;
the ABI does not include privileged system instructions, instead it
invokes system calls.

 Application Program Interface (API) - defines the set of instructions

the hardware was designed to execute and gives the application
access to the ISA; it includes HLL library calls which often invoke
system calls.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 7
 A1 Applications

API

Libraries A2

ABI
System calls

Operating System A3

ISA
System ISA User ISA
Hardware

Application Programming Interface, Application Binary Interface,

and Instruction Set Architecture . An application uses library
functions (A1), makes system calls (A2), and executes machine
instructions (A3).
Cloud Computing: Theory and Practice.
Dan C. Marinescu Chapter 5 8
Code portability

 Binaries created by a compiler for a specific ISA and a specific

operating systems are not portable.

 It is possible, though, to compile a HLL program for a virtual

machine (VM) environment where portable code is produced
and distributed and then converted by binary translators to the
ISA of the host system.
A dynamic binary translation converts blocks of guest
instructions from the portable code to the host instruction and
leads to a significant performance improvement, as such blocks
are cached and reused

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 9
 HLL code

Compiler front-end Compiler

Intermediate Portable
code code

Compiler back-end VM loader

Object code VM image

VM compiler/ VM compiler/
Loader
interpreter interpreter

Memory Memory Memory

image image ISA-1 image ISA-2

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 10
Virtual machine monitor (VMM / hypervisor)

 Partitions the resources of computer system into one or more virtual

machines (VMs). Allows several operating systems to run
concurrently on a single hardware platform.

 A VMM allows
 Multiple services to share the same platform.
 Live migration - the movement of a server from one platform to
another.
 System modification while maintaining backward compatibility
with the original system.
 Enforces isolation among the systems, thus security.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 11
VMM virtualizes the CPU and the memory
 A VMM
 Traps the privileged instructions executed by a guest OS and
enforces the correctness and safety of the operation.
 Traps interrupts and dispatches them to the individual guest
operating systems.
 Controls the virtual memory management.
 Maintains a shadow page table for each guest OS and replicates
any modification made by the guest OS in its own shadow page
table. This shadow page table points to the actual page frame
and it is used by the Memory Management Unit (MMU) for
dynamic address translation.
 Monitors the system performance and takes corrective actions to
avoid performance degradation. For example, the VMM may
swap out a Virtual Machine to avoid thrashing.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 12
Virtual machines (VMs)

 VM - isolated environment that appears to be a whole computer,

but actually only has access to a portion of the computer
resources.
 Process VM - a virtual platform created for an individual process
and destroyed once the process terminates.
 System VM - supports an operating system together with many
user processes.
 Traditional VM - supports multiple virtual machines and runs
directly on the hardware.
 Hybrid VM - shares the hardware with a host operating system and
supports multiple virtual machines.
 Hosted VM - runs under a host operating system.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 13
Traditional, hybrid, and hosted VMs
Process VMs System VMs

Same ISA Different ISA Same ISA Different ISA

Application Application

Multi Dynamic Guest Guest

Traditional W hole
program translators OS -1 OS -n
VM system VM

Binary
optimizers HLL VMs
Hybrid VM
Codesigned VM-1 VM-n
VM

Virtual Machine Monitor

Hosted VM

Hardware

(a) (b)

Application Application
Application
Guest OS -1 Guest OS -n
Application
Application
Application

VM-1 VM-n
Guest OS
Virtual Machine Monitor

Host OS VMM Host OS

Hardware Hardware

(c) (d)

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 14
 Cloud Computing: Theory and Practice.
Dan C. Marinescu Chapter 5 15
Performance and security isolation
 The run-time behavior of an application is affected by other
applications running concurrently on the same platform and
competing for CPU cycles, cache, main memory, disk and network
access. Thus, it is difficult to predict the completion time!

 Performance isolation - a critical condition for QoS guarantees in

shared computing environments.

 A VMM is a much simpler and better specified system than a

traditional operating system. Example - Xen has approximately
60,000 lines of code; Denali has only about half, 30,000.

 The security vulnerability of VMMs is considerably reduced as the

systems expose a much smaller number of privileged functions.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 16
Computer architecture and virtualization
 Conditions for efficient virtualization:
 A program running under the VMM should exhibit a behavior
essentially identical to that demonstrated when running on an
equivalent machine directly.
 The VMM should be in complete control of the virtualized resources.
 A statistically significant fraction of machine instructions must be
executed without the intervention of the VMM.
 Two classes of machine instructions:
 Sensitive - require special precautions at execution time:
 Control sensitive - instructions that attempt to change either the
memory allocation or the privileged mode.
 Mode sensitive - instructions whose behavior is different in the
privileged mode.
 Innocuous - not sensitive.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 17
Full virtualization and paravirtualization

 Full virtualization – a guest OS can run unchanged under the VMM

as if it was running directly on the hardware platform.
 Requires a virtualizable architecture.
 Examples: Vmware.

 Paravirtualization - a guest operating system is modified to use only

instructions that can be virtualized. Reasons for paravirtualization:
 Some aspects of the hardware cannot be virtualized.
 Improved performance.
 Present a simpler interface.
Examples: Xen, Denaly

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 18
Full virtualization and paravirtualization

Guest OS Guest OS
Hardware Hardware
abstraction abstraction
layer layer

Hypervisor Hypervisor

Hardware Hardware

(a) Full virtualization (b) Paravirtualization

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 19
Virtualization of x86 architecture
 Ring de-privileging - a VMMs forces the operating system and the
applications to run at a privilege level greater than 0.
 Ring aliasing - a guest OS is forced to run at a privilege level other
than that it was originally designed for.
 Address space compression - a VMM uses parts of the guest
address space to store several system data structures.
 Non-faulting access to privileged state - several store instructions
can only be executed at privileged level 0 because they operate on
data structures that control the CPU operation. They fail silently
when executed at a privilege level other than 0.
 Guest system calls which cause transitions to/from privilege level 0
must be emulated by the VMM.
 Interrupt virtualization - in response to a physical interrupt, the VMM
generates a ``virtual interrupt'' and delivers it later to the target guest
OS which can mask interrupts.
Cloud Computing: Theory and Practice.
Dan C. Marinescu Chapter 5 20
Virtualization of x86 architecture (cont’d)
 Access to hidden state - elements of the system state, e.g.,
descriptor caches for segment registers, are hidden; there is no
mechanism for saving and restoring the hidden components when
there is a context switch from one VM to another.
 Ring compression - paging and segmentation protect VMM code
from being overwritten by guest OS and applications. Systems
running in 64-bit mode can only use paging, but paging does not
distinguish between privilege levels 0, 1, and 2, thus the guest OS
must run at privilege level 3, the so called (0/3/3) mode. Privilege
levels 1 and 2 cannot be used thus, the name ring compression.
 The task-priority register is frequently used by a guest OS; the
VMM must protect the access to this register and trap all attempts
to access it. This can cause a significant performance degradation.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 21
VT-x, a major architectural enhancement

 Supports two modes of operations:

 VMX root - for VMM operations.
 VMX non-root - support a VM.

 The Virtual Machine Control Structure including host-state and

guest-state areas.
 VM entry - the processor state is loaded from the guest-state of the VM
scheduled to run; then the control is transferred from VMM to the VM.
 VM exit - saves the processor state in the guest-state area of the
running VM; then it loads the processor state from the host-state area,
finally transfers control to the VMM.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 22
 VT- x

Virtual-machine control structure

VM entry
host-state
VMX root VMX non-root
guest-state
VM exit

(a) (b)

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 23
VT-d, a new virtualization architecture

 I/O MMU virtualization gives VMs direct access to

peripheral devices.
 VT-d supports:
 DMA address remapping, address translation for device DMA
transfers.
 Interrupt remapping, isolation of device interrupts and VM
routing.
 I/O device assignment, the devices can be assigned by an
administrator to a VM in any configurations.
 Reliability features, it reports and records DMA and interrupt
errors that my otherwise corrupt memory and impact VM
isolation.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 24
Xen - a VMM based on paravirtualization
 The goal of the Cambridge group - design a VMM capable of scaling
to about 100 VMs running standard applications and services
without any modifications to the Application Binary Interface (ABI).
 Linux, Minix, NetBSD, FreeBSD, NetWare, and OZONE can operate
as paravirtualized Xen guest OS running on x86, x86-64, Itanium,
and ARM architectures.
 Xen domain - ensemble of address spaces hosting a guest OS and
applications running under the guest OS. Runs on a virtual CPU.
 Dom0 - dedicated to execution of Xen control functions and privileged
instructions.
 DomU - a user domain.

 Applications make system calls using hypercalls processed

by Xen; privileged instructions issued by a guest OS are
paravirtualized and must be validated by Xen.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 25
 Xen
Management
OS Application Application Application

Guest OS Guest OS Guest OS

Xen-aware
device drivers
Xen-aware Xen-aware Xen-aware
device drivers device drivers device drivers

Xen
Domain0 control Virtual x86 Virtual physical Virtual block
interface Virtual network
CPU memory devices

X86 hardware

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 26
Xen implementation on x86 architecture
 Xen runs at privilege Level 0, the guest OS at Level 1, and
applications at Level 3.
 The x86 architecture does not support either the tagging of TLB
entries or the software management of the TLB. Thus, address
space switching, when the VMM activates a different OS, requires a
complete TLB flush; this has a negative impact on the performance.
 Solution - load Xen in a 64 MB segment at the top of each address
space and delegate the management of hardware page tables to
the guest OS with minimal intervention from Xen. This region is not
accessible or re-mappable by the guest OS.
 Xen schedules individual domains using the Borrowed Virtual Time
(BVT) scheduling algorithm.
 A guest OS must register with Xen a description table with the
addresses of exception handlers for validation.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 27
Dom0 components
 XenStore – a Dom0 process.
 Supports a system-wide registry and naming service.
 Implemented as a hierarchical key-value storage.
 A watch function informs listeners of changes of the key in storage
they have subscribed to.
 Communicates with guest VMs via shared memory using Dom0
privileges.
 Toolstack - responsible for creating, destroying, and managing the
resources and privileges of VMs.
 To create a new VM, a user provides a configuration file describing
memory and CPU allocations and device configurations.
 Toolstack parses this file and writes this information in XenStore.
 Takes advantage of Dom0 privileges to map guest memory, to load a
kernel and virtual BIOS and to set up initial communication channels
with XenStore and with the virtual console when a new VM is created.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 28
Strategies for virtual memory management, CPU multiplexing, and
I/O devices

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 29
Xen abstractions for networking and I/O
 Each domain has one or more Virtual Network Interfaces (VIFs)
which support the functionality of a network interface card. A VIF is
attached to a Virtual Firewall-Router (VFR).
 Split drivers have a front-end in the DomU and the back-end in
Dom0; the two communicate via a ring in shared memory.
 Ring - a circular queue of descriptors allocated by a domain and
accessible within Xen. Descriptors do not contain data, the data
buffers are allocated off-band by the guest OS.
 Two rings of buffer descriptors, one for packet sending and one for
packet receiving, are supported.
 To transmit a packet:
 a guest OS enqueues a buffer descriptor to the send ring,
 then Xen copies the descriptor and checks safety,
 copies only the packet header, not the payload, and
 executes the matching rules.
Cloud Computing: Theory and Practice.
Dan C. Marinescu Chapter 5 30
 I/O channel
Driver domain Guest domain

Bridge
Frontend
Backend

Network
interface

Event channel

XEN

NIC
(a)

Request queue

Consumer Request Producer Request

(private pointer in Xen) (shared pointer updated
by the guest OS)

Outstanding Unused
descriptors descriptors

Consumer Response
Producer Response
(private pointer maintained by
(shared pointer updated
Response queue the guest OS)
by Xen)
(b)

Xen zero-copy semantics for data transfer using I/O rings. (a) The communication
between a guest domain and the driver domain over an I/O and an event channel;
NIC is the Network Interface Controller. (b) the circular ring of buffers.
Cloud Computing: Theory and Practice.
Dan C. Marinescu Chapter 5 31
Xen 2.0
 Optimization of:
 Virtual interface - takes advantage of the capabilities of some
physical NICs, such as checksum offload.
 I/O channel - rather than copying a data buffer holding a packet,
each packet is allocated in a new page and then the physical
page containing the packet is re-mapped into the target domain.
 Virtual memory - takes advantage of the superpage and global
page mapping hardware on Pentium and Pentium Pro
processors. A superpage entry covers 1,024 pages of physical
memory and the address translation mechanism maps a set of
contiguous pages to a set of contiguous physical pages. This
helps reduce the number of TLB misses.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 32
 Driver domain Guest domain Driver domain Guest domain

Bridge Bridge

Offload
I/O
Driver I/O
channel channel High Level
NIC Backend Virtual NIC Backend Virtual
Driver Interface Interface Driver Interface
Interface

Physical Xen VMM Physical Xen VMM

NIC NIC

(a) (b)

Xen network architecture .(a) The original architecture;

(b) The optimized architecture

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 33
 A comparison of send and receive data rates for a native Linux system, the Xen
driver domain, an original Xen guest domain, and an optimized Xen guest domain.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 34
Performance comparison of virtual machines
 Compare the performance of Xen and OpenVZwith, a standard
operating system, a plain vanilla Linux.
 The questions examined are:
 How the performance scales up with the load?
 What is the impact of a mix of applications?
 What are the implications of the load assignment on individual
servers?
 The main conclusions:
 The virtualization overhead of Xen is considerably higher than that of
OpenVZ and that this is due primarily to L2-cache misses.
 The performance degradation when the workload increases is also
noticeable for Xen.
 Hosting multiple tiers of the same application on the same server is
not an optimal solution.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 35
 Linux OpenVZ Xen

Web Web Web

server server server

MySQL MySQL MySQL

server server server

(a)

Linux OpenVZ Xen

Web MySQL Web MySQL Web MySQL

server server server server server server

(b)

Linux OpenVZ Xen

Web MySQL Web MySQL Web MySQL
server server server server server server
Web MySQL Web MySQL Web MySQL
server server server server server server
Web MySQL Web MySQL Web MySQL
server server server server server server
Web MySQL Web MySQL Web MySQL
server server server server server server

(c)

The setup for the performance comparison of a native Linux system with OpenVZ, and
the Xen systems. The applications are a web server and a MySQL database server. (a)
The first experiment, the web and the DB, share a single system; (b) The second
experiment, the web and the DB, run on two different systems; (c) The third experiment,
the web and the DB, run on two different systems and each has four instances.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 36
The darker side of virtualization
 In a layered structure, a defense mechanism at some layer can be
disabled by malware running at a layer below it.
 It is feasible to insert a rogue VMM, a Virtual-Machine Based Rootkit
(VMBR) between the physical hardware and an operating system.
 Rootkit - malware with a privileged access to a system.
 The VMBR can enable a separate malicious OS to run
surreptitiously and make this malicious OS invisible to the guest OS
and to the application running under it.
 Under the protection of the VMBR, the malicious OS could:
 observe the data, the events, or the state of the target system.
 run services, such as spam relays or distributed denial-of-service
attacks.
 interfere with the application.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 37
 Application

Application
Malicious Guest OS
OS
Operating
Malicious system (OS)
OS Virtual machine monitor

Virtual machine based rootkit Virtual machine based rootkit

Hardware Hardware

(a) (b)
The insertion of a Virtual-Machine Based Rootkit (VMBR) as the lowest
layer of the software stack running on the physical hardware; (a) below an
operating system; (b) below a legitimate virtual machine monitor. The
VMBR enables a malicious OS to run surreptitiously and makes it invisible
to the genuine or the guest OS and to the application.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 38
The features of the SFI for the Native Client on the x86-32, x86-64 , and ARM.

Cloud Computing: Theory and Practice.

Dan C. Marinescu Chapter 5 39