virtualization and hypervisors

 Detailed study of a virtual technology, its
types, advantages and disadvantages.
 Study about Hypervisors or Virtual
Machine Monitors ,its types and
examples.
 Choose hypervisors for selection on the
bases of its audience.
 Compare hypervisors on technical
differences,

 4 tools to compare i.e. XEN, KVM, Vmware and VirtualBox.
Initially, we will study all the features of the tools and hence
compare them on technical ground. After comparing
technically we will compare them on benchmarks such as cpu
speed, network speed, i/o speed, etc. further we will compare
on real worls scenarios such as ftp server, sql server,
decompression/compression.
 BENEFIT- As a result we will get the tool best for the virtualization
in given situation. Some of the comparing attributes would be os
support of tool, security issues, cpu speed, network speed, i/o
speed, least rtt(round trip time), etc.
 A number of research works exist in the development of
Hypervisor. However, the search for the ideal hypervisor
continues.

Virtualization allows multiple operating system instances to
run concurrently on a single computer; it is a means of
separating hardware from a single operating system.

 Coexistence of Operating Systems on the
same machine
 Protection
 Operating System research
 Software testing and runtime debugging
 Optimization of hardware utilization
 Job migration
 Virtual storage
 Back Up an Entire Operating System
 AND MANY MORE…..

1. CPU virtualization
1.1. full virtualization
1.2 para-virtualization
1.3 hardware assisted virtualization
2. Memory virtualization
3. Device and I/O virtualization

Unlike Full virtualization, Paravirtualization involves modifying the OS
kernel to replace nonvirtualizable instructions with hypercalls that
communicate directly with the virtualization layer hypervisor.

 Beyond CPU virtualization, This involves sharing the
physical system memory and dynamically
allocating it to virtual machines.
 The operating system keeps mappings of virtual
page numbers to physical page numbers stored
in page tables. All modern x86 CPUs include a
memory management unit (MMU) and a
translation lookaside buffer (TLB) to optimize virtual
memory performance.

 The final component required beyond CPU
and memory virtualization is device and I/O
virtualization. This involves managing the
routing of I/O requests between virtual
devices and the shared physical hardware.
The hypervisor
virtualizes the
physical hardware
and presents each
virtual machine with
a standardized set
of virtual devices as
seen in Figure .

 A hypervisor or virtual machine
monitor (VMM) is a piece of computer
software, firmware or hardware that
creates and runs virtual machines.
 A computer on which a hypervisor is
running one or more virtual machines is
defined as a host machine.
 Each virtual machine is called a guest
machine.

 Type 1 hypervisor is installed directly on
bare-metal hardware, it doesn't require
an additional OS, it is the OS, even it it is
a light or minimal OS.
 EX. Kvm and Xen
 Advantages: System is thin, the hypervisor
has direct access to the HW, higher
density hardware.
 Disadvantages: Really, Really large VMs
are not supported, HW should support
virtualization technology, costlier and
Really bad console interface.

 Type 2 is more of an application installed on
an operating system and not directly on the
bare-metal.
 EX. VirtualBox and Vmware Workstation
 Advantages: Run on a greater array of HW
because the underlying Host OS is
controlling HW access, Easy user interface,
Data can be secured on the desktop .
 Disadvantages: Decreased security, Loss of
Centralized Management, Lower VM
Density, Cannot support as many VMs are
the first type.

virtualization and hypervisors

•Primary — The single main
hypervisor used as the standard for
virtualizing their servers. The
survey allowed only one response in
this category.
•Also Use — Respondents were
asked to report other hypervisors
deployed in their datacenter. There
was no limit to the number of these
responses and many organizations
had more than one secondary
hypervisor.
•Plan to Stop —This was to capture
which hypervisors organizations
have currently deployed but planned
to cease using in the near future.
•Evaluating — Aberdeen asked if
there were hypervisors that are
being evaluated or considered for
future datacenter deployment.

 Vmware(type 2) – Bhavya k. Saxena
 XEN (type 1) – Gaurav suri
 KVM (type 1)– Rishi Golyan
 VirtualBox (type 2) – Pupul Saxena

 Supports bridging existing host network
adapters and share physical disk
drives and USB devices with a virtual machine.
 It can simulate disk drives. It can mount an
existing ISO image file into a virtual optical disc
drive so that the virtual machine sees it as a
real one. Likewise, virtual hard disk drives are
made via .vmdk files
 VMware Workstation can save the state of a
virtual machine (a "snapshot") at any instant.
These snapshots can later be restored,
effectively returning the virtual machine to the
saved state.

 Infrastructure : Vmware Virtual machine file
system(VMFS), Vmware Virtual Symmetric Multi
processing(SMP), Virtual Infrestucture web access,
Vmware Vmotion, Vmware Distributed Resource
Scheduler
 Storage And Arrays:Fiber Channel SAN arrays, iSCSI
SAN arrays and NAS arrays are widely-used storage
technologies supported by Vmware Infrastructure to
meet different data center storage needs
 Ip Networks: Each computing server can have
multiple gigabit Ethernet network interface
cards(NICs) to provide high bandwidth and reliable
networking to the entire data center

proscons
 Encrypts a virtual machine as a way to keep it
from being used by unauthorized personnel
 Each VM can now support up to 16 virtual
CPUs and 16 cores per virtual CPU, up to
64GB of RAM
 Can configure up to 20 virtual networks
within a single instance of Workstation.
 Availability of hardware-specific features to
guests is constrained by the actual machine
 VMware is overpriced.
 Availability, reliability, Complexity issue.
 Reduced application performance
 Hyper-V and XenServer are better
alternatives
 Hardware compatibility
 Application compatibility

 Portability. VirtualBox runs on a large number of 32-bit and 64-bit host
operating systems.
 No hardware virtualization required.
 Guest Additions. After installing the Guest Additions, a virtual machine
will support automatic adjustment of video resolutions, seamless
windows, accelerated 3D graphics and more. Hence, improving the
performance of the guest OS and providing additional integration and
communication with the host system.
 Great hardware support.
 Guest multiprocessing- VirtualBox can present up to 32 virtual CPUs to
each virtual machine, irrespective of how many CPU cores are
physically present on your host.
 USB device support, Hardware compatibility, Full ACPI support, PXE
Network boot, Built-in iSCSI support.
 Multigeneration branched snapshots.
 Clean architecture; unprecedented modularity.
 Remote machine display.
 Extensible RDP authentication, USB over RDP

 Offers built-in GUI-based wizard for cloning a VM.
 Offers built-in VM snapshot, and supports multiple
snapshots.
 Supports multiple types of disk image (e.g., vdi, vmdk, vhd,
hdd, qed, qcow).
 Supports multiple virtual monitors for a guest VM. Multiple
virtual monitors belonging to a VM can be displayed in
multiple windows of host machine, or via multiple physical
monitors.
 Can be installed on OS X, Windows 7, Windows 2008
Server, Solaris, OpenSolaris, FreeBSD host operating
systems.
 More frequent minor version upgrades.
 Comprehensive built-in command-line interface for
creating and running VMs (e.g., VBoxManage).

 OVF localization (multiple languages in one OVF file) is not
yet supported.
 Cannot access NATed guest VMs from host machine
without enabling port forwarding on host machine.
 NATed guest VMs are isolated from one other, and cannot
communicate with one another.
 Bridged networking or host-only networking is required if
guest VMs need to communicate with one another on the
same Layer-2 network.
 Can add up to eight vNICs to a guest VM, while VMware
Player can create up to ten vNICs per guest VM. You can
configure only four vNICs via VirtualBox GUI. To add more
than four vNICs, you need to use VirtualBox CLI.
 Does not support nested hardware-assisted virtualization.
 No support for USB 3.0.

 EFI (extensible Firmware Interface) support for hypervisor.
Allows Xen to boot on machines which use EFI rather than a
traditional BIOS
 Support up to 4095 Host CPUs for 64 bit h/v (from 256)
 Support for dom0 kernels compressed with xz
 Per-device interrupt remapping (increases scalability)
 Support for PVHVM guest direct physical IRQ injection
(improves performance for PCI passthrough to Linux Guests)
 Multiple PCI segment support
 Lots of XSM / Flask fixes (security)
 AMD SVM "DecodeAssist" support (AMD CPU feature that
avoids emulation and increases performance)
 Credit Scheduler tuning parameters:
› sched_ratelimit_us
› tslice_ms

 AMD OSVW (OS Visible Workarounds): Disables OS
workarounds for hardware errata which are not necessary to
workaround in guests because it is handled at the host level.
 Early boot time CPU microcode patching. Xen can supplied
with the microcode image by the bootloader and load it
early rather than relying on the domain 0 kernel to supply it
later.
 Improvements to paging and sharing, enabling higher VM
density for VDI use-cases
› Heavily reworked page sharing. This remains a tech preview though due
to limited tools support.
 Enhanced memaccess interfaces allowing increased
introspection of guest memory by tools.
 Initial support for nested virtualisation. This allows HVM guests
access to hardware virtualisation features such that they can
run their own hypervisor.

proscons
 It is more modernized than KVM, user friendly
and can run without hardware support
 Support for multiple hardware platforms
 High security isolation features, independent
of operating systems
 Offers paravirtualization and hardware
assisted virtualization
 A different, user-friendly interface
 It is not included in Linux except in recent
versions
 It has minimal power management
features
 It has a different dom0 structure
increasing overload on CPU.
 More complex than other hypervisors

 NPT/EPT support (server boost)
 KSM (share memory with COW)
 Disk image cloning, sharing, snapshot
 Ballooning
 Live migration (nfs as shared storage)
 Save and restore VM
 Virtio paravirtualization
 PCI-passthrough VT-D/IOMMU support

 Each virtual CPU appears as a regular
Linux process
 Emulation is handle by a modified
version of QEMU
 Linux as a VMM
 Resource management
 The KVM control interface
 Emulation of hardware

proscons
 It is free.
 It is a part of Linux
 Powerful CPU virtualization on Intel and AMD
platforms, leverages hardware virtualization and
Linux kernel, memory management, I/O capabilities
 Real time scheduling
 Powerful GUI
 Powerful command line scripting tools for better
productivity
 Leverages security, kernel capabilities and memory
managemnt of Linux
 A bit dated
 Difficult to master for novices and those used
to GUI based usage
 Very slow when CPU does not support
virtualization and when it works in QEMU
mode
 Heavy I/O overhead
 Lack of flexibility in device assignment
 Issues in memory and power management

1. Host operating system support(virtualbox
is better)
2. Ease of editing virtual machine’s
configuration (virtualbox is better)
3. Usb support (vmware is better)
4. Range of virtual hard disks (virtualbox is
better)
5. Remote connections (virtualbox is better)
6. VM cloning (virtualbox)

7. Ease of boot (vmware is better)
8. USB over RDP (virtualbox is better)
9. "Teleportation" migration
functionality.(virtualbox is better)
10. Command-line options.(virtualbox is
better)
11. Graphics(VMware is better)
12. Ovf support(vmware is better)

1. HOST OS - KVM isn't an option on older CPUs made before
the virtualization extensions were developed, and it rules out
newer CPUs (like Intel's Atom CPUs) that don't include
virtualization extensions.
2. Red Hat drops XEN and anoints KVM
3. Market of KVM and XEN - If you're going with RHEL over
the long haul, bank on KVM. If you're running on
Amazon's EC2, you're already using Xen, and so on
4. Operating system overhead - Xen is not burdened with any
operating system overhead that is unrelated to processing a series
of guests on a given machine
5. Security - Xen ensures a high level of security via a variety of
methods/features: guest isolation,priviliged access, small code
base and operating system seperation

6. Maturity - The Xen hypervisor has been available for enterprise
deployment since 2004 and is the first open source hypervisor to
successfully be deployed by industry leading Linux vendors
7. Scheduling -Xen uses its own kernel for thread scheduling and
dispatching virtual machines, while KVM, accepted into mainline
Linux kernel sources, uses that kernel for these operations
8. Ease of use - KVM is generally considered easier to
configure and operate given it is just a single module
that you load in the Linux kernel.
9. Memory page sharing -XEN doesn't implement memory
page sharing and KVM does it very efficiently.
10. Lack of Dom0 in kvm -KVM introduces many
performance benefits, such as less I/O latency due to
lack of Dom0
11. I/O and network operations –
Kvm- VM -> Virtio -> CentOS with KVM
XEN- VM -> PV-OPS Driver -> Dom0 -> Xen

1. CPU speed
2. Cache and Memory speed
3. i/o Disk access speed
4. Network Speed
5. Cryptography speed

VMware VirtualBox KVM Xen
SSE yes yes Yes yes
SSE2 yes yes Yes yes
SSE3 /
SSSE3
yes / yes yes / no yes / no yes / yes
SSE4 yes no no yes
1. VMware takes the lead in both ALU
and FPU performance, but it is the
integer speed which shows the
higher advantage.
2. Only VMware and Xen present to
the guest OS the SSSE3 and newer
extensions.
3. The net result is that VirtualBox and
KVM are on par with VMware, while
Xen show some problems giving a
considerably lower result.

1. While VMware, VirtualBox
and KVM give us quite similar
in results, Xen is behind the
competitors. This can be a
results of a more-heavy
hypervisor or, more probably,
of the “double
encapsulation” of the guest
system
2. The story is reversed now:
Xen is slight faster, and least
VirtualBox.
3. It seems that Xen do a
very good use of the nested
page table feature.

1.KVM is, by much, the
slower hypervisor in
the read test;
VMware is best.
2.To us, the KVM results
are due to very poor
caching and great
I/O overhead.

1. Watching the write
test, we have a very
different picture: the
faster virtual machine
is VirtualBox
2. To us, seems that
VirtualBox use a
write-back cache
algorithm, while the
others use a write-
throught policy.
3. greater speed for
VirtualBox, but also a
greater risk of data
loss
1 2 4 8 16 32
XEN 410,28 367,6 370,37 374,77 376,66 376,78
KVM 108,52 107,22 105,81 109,3 107,17 95,8
VBOX 2022,38 8548,59 4756,33 3162,37 1120,3 2044,4
VMWARE 216,63 267,01 332,63 361,53 359,8 393,46

 ping -f 192.168.122.166 -c 25000 -s 1024
1. The results are quite close, with Xen
only slight slower
2. We can see that KVM was the best
performing virtual machine, with a
privileged time dominated by IRQ
servicing time: this means that the
network-related syscalls executed
quite fast on KVM. On contrary, Xen
has a very high privilege time but a
lower IRQ time: the network-related
syscalls executed slowly. VMware as a
extraordinary low IRQ time, probably a
courtesy of its paravirtualized network
driver, while VirtualBox is overall a
decent performer.

1. Installation time
2. Web server test
3. Mysql performance
4. FTP server test
5. File compression/ Decompression test

 VMware and Xen are very
closely matched, while KVM is
the real loser: it took over 30
minutes to complete, a 5-fold
increase compared to
VirtualBox!

1. Xen is really slow, the other
hypervisors perform quite
similarly here. As Apache
creates a new thread for
each new connection, it is
entirely possible that the real
culprit here is a very low
speed in creating new
threads.
2. VirtualBox seems to be the
best user-level hypervisior, but
it lag behind in kernel-level
CPU time. At the other end,
VMware has great kernel-
level CPU time but the worst
user-level time.
Xen is the slowest
machine, while VMware
is the fastest.

1. Xen was the slowest machine, while VirtualBox
was the fastest
2. Xen is again the slowest hypervisor, with a very
great gap from the other. Well, it seems that we
were right: MySQL is another well-threaded
program, and Xen results are very low. On the other
hand, KVM is a little faster that the others.
3. We see a very great dominance of privileged
(kernel) time. It means that the system spend the
most time on syscalls or IRQ servicing routines.

1. Xen-best- download vmware-best-upload
2. The results are not so homogeneous now:
Xen is the least efficient hypervisor here
showing syscall routines execute very slowly.
3. The download disk activity is so small thanks
to the guest side caching. VMware is the
most efficient machine, Xen is the least.

1. Apart Xen, which is slower, VMware, VirtualBox
and KVM are quite paired each other.
2. KVM is very fast to both compressing and
decompressing.
3. compression test, which is more CPU intensive,
KVM and VirtualBox both have a quite strong
advantage. In the decompression test, which is
generally disk bound, we see that KVM lost the
crown probably as a result of its no-so-quick disk
subsystem, while Xen and VMware are slightly
better.

Test Best Worst
Cpu speed Vmware Xen
Cache subsystem Vmware Xen
Memory subsystem Xen Vbox
I/O speed Vbox KVM
Network speed KVM Xen
Cryptographic speed Vmware, KVM, Vbox Xen
Installation time Vbox KVM
Web server test Vbox followed by vmware Xen
Sql server test KVM Xen
FTP server test Vmware Xen
Compression/ Decompression speed KVM / Vmware Xen / Vbox

virtualization and hypervisors

Recommended

More Related Content

What's hot (20)

Viewers also liked (14)

Similar to virtualization and hypervisors (20)

Recently uploaded (20)

virtualization and hypervisors

Editor's Notes