Zhou, F.F., Ma, R.H., Li, J., Chen, L.X., Qiu,
W.D. and Guan, H.B., 2016. Optimizations for high performance network
virtualization. Journal of Computer Science and Technology, 31(1), pp.107-116.
Van, O.H. and Gray, J., Glue Networks, Inc.,
2016. Systems and methods for determining endpoint configurations for endpoints
of a virtual private network (VPN) and deploying the configurations to the
endpoints. U.S. Patent 9,319,300.
Sun, Y., Xu, M., Lian, J.J.R. and Shieh, C.Y.M.,
Varmour Networks, Inc., 2016. System and method for dynamic security insertion
in network virtualization. U.S. Patent 9,258,275.
R., A Security Model for Virtual Infrastructure in the Cloud.
Mijumbi, R., Serrat,
J., Gorricho, J.L., Bouten, N., De Turck, F. and Boutaba, R., 2016. Network
function virtualization: State-of-the-art and research challenges. IEEE
Communications Surveys & Tutorials, 18(1), pp.236-262.
A., Schrepfer, S., Bocek, T. and Stiller, B., 2016, April. Overlay network
measurements with distribution evolution and geographical visualization. In
NOMS (Vol. 2016, pp. 222-230).
Kratzke, N., 2017. About microservices,
containers and their underestimated impact on network performance. arXiv
Harmening, J.T., 2017. Virtual private networks.
In Computer and Information Security Handbook (Third Edition) (pp. 843-856).
F., Openvpn Technologies, Inc. and OPENVPN TECHNOLOGIES Inc, 2017. Private
tunnel network. U.S. Patent 9,699,135.
Denning, P.J., The Science of Computing.
American Scientist, 77, pp.216-219.
A., Basta, A., Reisslein, M. and Kellerer, W., 2016. Survey on network
virtualization hypervisors for software defined networking. IEEE Communications
Surveys & Tutorials, 18(1), pp.655-685.
S., Trach, B., Gregor, F., Knauth, T., Martin, A., Priebe, C., Lind, J.,
Muthukumaran, D., O’Keeffe, D., Stillwell, M. and Goltzsche, D., 2016,
November. SCONE: Secure Linux Containers with Intel SGX. In OSDI (pp. 689-703).
Anderson, J., Hu, H., Agarwal, U., Lowery, C.,
Li, H. and Apon, A., 2016, February. Performance considerations of network
functions virtualization using containers. In Computing, Networking and
Communications (ICNC), 2016 International Conference on (pp. 1-7). IEEE.
Ahmed, F., Butt, Z.U. and Siddiqui, U.A., 2016.
MPLS based VPN Implementation in a Corporate Environment. Journal of
Information Technology & Software Engineering, 6(5), pp.1-7.
In conclusion it can be stated that this report can
successfully provide a brief conceptual model of NVE. Researchers can get
guidance from the principles of NVE while deigning a VN and developing the
algorithm for further studies. This also reflects on the goals of VN that lead
to the realisation of the importance of further studies on the VN in global
context of communication networking.
Legacy support- it is nothing but
the implementation of backward computability in new VN versions. For example
the IPV6 can become more faster in case of implementing it in IPV4.
Stability and Convergence- NVE can
be destabilised by errors and miss-configurations of underlying physical network and that can
be maintained by ensuring the stability of NVE by implementing proper
conversing scope V.
Scalability- in order to increase
the scalability the number of co-existing VNs must be increased but without
affecting the performance.
Manageability- the network
management task can be modularized by separating the SPs from InPs and in order
to do that accountability must be introduced to each and every layer of VN. From
example deploying sources is a very difficult task as there exists lack of
consensus among the ISPs.
Flexibility- this is the key point
that has to be maintain while making a network design. The topology, routing
and function forwarding must be customisable for the end of each SPs II.
3.3. Design goals
4 is a clear example of Revisitation in the VN2.
This helps in hosting of multiple virtual nodes of a single VN by
In fig 4. All the constraints
existing in InP2 will be transmitted to VN2 from VN1.
architectural attributes of parents can be inherited by Child NVs in NVE that
initiate an automated translation of the parent’s constraint to the Child VNs
Recursion- This principle is
associated with parent-child relationship that is generated during the spawning
of multiple VN from one to another in order to create network hierarchy. In fig
4 the topmost VN is created by SP0, the used physical network is provided by
InP0 and is allotted to SP1. Then leased by InP1 and they created the parent-child
Coexistence- it is termed as the
defining characteristics of NVE according to which different service providers
can co-exist and can use the underlying physical networks provided by multiple
InPs III. E.g. VN1 and VN2 in fig 4.
VN2- this is created by
combining the recourse from LnP1 and a child VN from Sp1. The customers of this
VN are U1 and U3.
VN1- physical resources
of this VN is managed by Inp1 and InP2. There two customers are U1 and U2. The
service provide by them in end-to- end in nature.
In figure 4 the creation of two
VN- VN1 and VN2 has been described. Two VN is from two different service
provides named SP1 and SP2 respectively.
The basic unit of NVE is the VN.
Each and every VN is hosted by particular physical node and a range of Virtual
links spans over the path of these virtual nodes I. There exist single SPs to
manage each and every VN but it must be ensured that all the physical resources
are aggregated from multiple Inps.
3.1 Architecture overview
The role of broker NV economy is pivotal in nature. They
bridge the communication between Inps, SPs and end-users. All the leases are
bought by brokers from InPs and VNs are created then these VNs are sold to the
interested customers according to their demand of services VI.
Figure 3. Relationship between players
(ECAs)- Between the
End user connectivity agreements
Service level agreements (SLAs)-
between service provider and customers.
In case of multiple virtual networks
an end user is allowed to connect with multiple providers with respect to his
required type of service, otherwise in case of normal virtual network the end
user is similar to the end-user of existing Internet service. The agreements
required for End-users are
The role of service provider is
to crate VN and deploy various customized protocol by leasing various recourses
from different facilities provides. In some cases they also do some programming
that are required for allocation of network resources for building of
end-to-end network services. Network provisioning agreements (NPAs) regulates
the relationship between InPs and SPs. In case of peering relationships the
required agreement is interconnection agreements (SIAs) IX.
2.2 Service Providers (SP)
Provider are the InPs engaged in
connecting the customer premise equipments (CPEs) while facilities providers offers different types of networking
technologies like optical fibre, satellite etc.
The key roles of InPs are to manage the operation of entire
physical infrastructure. Usually they are distinguished by their provided
resource quality, delegated freedom to their customers etc. In order to
maintain an end-to-end physical infrastructure collaboration is made between
multiple InPs in accordance with interconnection agreements (IIAs) XIV.
Figure2- Hierarchy of Roles
2.1 Infrastructure provide (InP)
VN business model is quite
different from that of traditional business model. The key players of this
model are described below,
The ability to create virtual networks decoupled from a
range of underlying network hardware is termed as network
Virtualisation (NV). The key aim of NV is to provide better integration with
increasingly virtual environment. In one word NV is an intelligent abstraction
networking platform and resources XII.
III. NETWORK VIRTUALIZATION ENVIRONMENT (NVE)
The radically different concepts can’t be
supported by this model as most of the overlays exist in application layers.
It doesn’t contain any holistic view as it is
mainly used for narrow fixes.
The Drawbacks of
Ensure the QoS guarantee
Provide Multi-casting features
Enables performance and availability of network routing
The key benefits of
Overlays VPN can be implemented in various type of switched WAN Layer 2
topologies such as Frame Relay, SMDS or ATM etc. Overlay networking can also be
used in Generic Router Encapsulation (GRE) and IPSec encryption V.
The degree of maximum bandwidth availability (Peak-Information-Rate or
PIR) and bandwidth guarantee on certain VC (Committed-Information-Rate or CIR) measures
the level of the QoS guarantee in the overlay. The statistical nature of Layer
2 service provide the CIR but it also influenced by RSP’s overbooking strategy
which reveals that the CIR is not always guaranteed and there arise a provision
of Minimum-Information-Rate ( MIR).
A router-to-router communication is established from customer side among
all Customer-premises-Equipment (CPE) devices. A continuous exchange of routing-Protocol
occurs within the customer devices providing any knowledge of the internal
structure of these devices to RSP IX.
overlay network a customers is provided with a set of VC lines (Lease-lines)
from the Routing service provider (RSP) site. The VC lines are of two types-
PVCs (Constantly available) and SVCs (available on demand). The topology has
been described in the figure 1
A range of network abstraction layers can be created using Overlay
Network. These layers are used for
running multiple discrete Virtual networks these usually consist of additional
security applications. Multiple secure overlays can be created using additional
software along with the already existing network hardware infrastructure. The
two end point of this overlay can be either a physical user or a network port
(e.g. Cloud User). A specific identification tag like a phone number that
become required during the identification and verification of the end-users is
incorporated with the overlay network and the entire virtual connection is established
on the basic of this tag or number XIII.
The key function of VRF includes the creation of separate and
private IP routing tables within one or more connected routers. VRF ensure that
the different IP routing table having same IP address within the connected
router will never overlap with each other. In most of the cases VRF works in
association with MPLS
Virtual Route Forward
Protocol Security ) VPN- this is known as the most secure means of creating VPN with the addition of security
features to the VPN packets.
2 Tunnel Protocol) VPN- the key feature of this VPN protocol is that it
utilizes encryption protocol of the tunnel but never provide any kind of
encryption by its own. Actually it’s a combination of L2F and PPTP VPN.
(Point-to-point Tunnelling Protocol ) VPN- this is the most popular Microsoft created VPN
protocol used mainly for the purpose of accessing geo-restricted content over
internet. This works on TCP and GRE port ConfigurationI.
VPN- This used for the purpose of accessing remote user. They key a feature
of this VPN is that no specialised client software installation required for
end user systems.
VPN- A direct, protected and unshared connection is between to end users is
established by this type of VPN. The VPN can be created either on intranet
based (e.g. organisation’s property network) or extranet based (e.g. external
partner network) V.
VPN can be defined as a protected tunnel through out a public network
that connects two or more devices X. According to the functionality and
protection level a range of variety lies in the field of VPN few examples are
Virtual Private Network (VPN)
Interfaces are usually derived from such types of software that does not
have any kind of physical properties by their own. Loopback is an example of
Virtual interface used for the signal testing purpose during routing. Other
examples are SVI, null, tunnels etc VI.
The first virtual circuit was X.25 and latter Frame Relay and ATM merit
was invented. This circuit used to work legacy protocols carrying either Switch
Virtual Circuit (SVC) or Permanent Virtual Circuit (PVC). These protocols have
been replaced by WAN virtualisation in modern days IV.
In the late 60″s the first router was developed by BBN, in 1980 a
multiprotocol router was developed by Bill Yeager. In 1993 the first successful
enterprise level multiprotocol router having Advance Gateway Server (AGS) was
invented by CISCO XI.
A Router is Born
The concept of virtualisation raised form the far back of 60’s. The
initial concept used to see the virtualisation as a method of logical system
resource division within various types of application in mainframes. Since then
it has been broaden its branches to variety of computing concepts like database
virtualisation, storage virtualisation, network virtualisation, software
virtualisation etc. Our focus will be concentrated on the network
virtualisation. The concept of network virtualisation stared from late 1981 by
Dr. David Sincoskie during an experiment regarding the segmentation of voice
over Ethernet broadcast network IX. It must be mentioned in this context that the Spanning Tree protocol
had been invented already and the problem related to fault tolerance and
redundant path was already solved hence the work of Dr. David was on the
further progress of networking but it was unsuccessful until the adaptation of
802.1 D in the IEEE in the year 1990. On 19th century switched
network use to dominate and the hubs and repeaters used to enjoy the higher
level of reliance. But now a day’s this hardware has become outdated and has
been replaced by bridges, Virtual LANs VII.
This entire report will be focused on past and present state of network
variations along with the identification of its future prospects.
The concept of network virtualisation is quite different for purists and
pluralist. From purist view network virtualisation seems to be the evaluation
of new network architecture while according to pluralist’s view it is
considered as the fundamental attributes of network architecture itself and the
ossifying forces of present internet can be extenuated by this virtualisation. The
network innovation can also be stimulated via implementation of diverse network
architectures III. But the introduction and implementation of diversity in
network is a very critical task as it must be carried out by maintaining a
range of policies that must be generated
from various available mechanism created for well-tested on the basis of
principle from available literatures. In the field of network virtualisation
the Internet service providers (ISPs) plays vital roles and their roles are
ever changing in nature. They have been sub-divided into two categories named
infrastructure providers and service providers. The physical infrastructures
are managed by the former categories while the latter is responsible for creating
and maintain the recourses related to virtual network.
Internet is successfully remodelling the way of accessing and exchanging
information in the modern tech-savvy world. Network virtualisation can be
defined as the process of software and hardware networking network resource
combination that converts the network functionality into a single software base
administrative entity. Modern human habitation has become totally dependent on
networking technology. The networking technology is ever changing is nature and
from past few decades the internet architecture has been remodelled and
modified in a continuous manner. The support and options of networking technologies
is becoming wider day by day II. The current technology on with the
networking technology runs is stunningly updated but its popularity lies on the
scope of its future growth. Due to its multi-provider nature the networking technology
as well as virtualisation always stays under a harsh competitive stress, hence
for the sack of sustainability continuous adaptation of fresh architectural
model and remodelling or modification of existing architecture become
Keywords: Cloud computing, Inter-networking,
Since past three decades the communication network has been changed a lot. The
new trend shows that the networking is becoming more software base and the use
of virtualisation in this context is getting more and more significance with ongoing
time. The popularity of Virtualisation is in high rise as this is capable of
providing a high performance, flexible and low cost software base network.
Moreover this type of networking service effectively broadens the networking
capabilities options in both inter-network and cloud performance hence network
virtualisation is termed as the ‘next generation penicillin’ for future inter-networking
prototype and cloud computing. This article is consists of some past findings
along with the modern status and future prospect of network virtualisation
Technology in Linux.