17-09-2016, 11:36 AM
1455093754-06553676.pdf (Size: 414.2 KB / Downloads: 3)
ABSTRACT
Cloud services are exploding, and organizations
are converging their data centers in order
to take advantage of the predictability, continuity,
and quality of service delivered by virtualization
technologies. In parallel, energy-efficient
and high-security networking is of increasing
importance. Network operators, and service and
product providers require a new network solution
to efficiently tackle the increasing demands
of this changing network landscape. Softwaredefined
networking has emerged as an efficient
network technology capable of supporting the
dynamic nature of future network functions and
intelligent applications while lowering operating
costs through simplified hardware, software, and
management. In this article, the question of how
to achieve a successful carrier grade network
with software-defined networking is raised. Specific
focus is placed on the challenges of network
performance, scalability, security, and interoperability
with the proposal of potential solution
directions.
INTRODUCTION: WHAT IS
SOFTWARE-DEFINED NETWORKING?
Network configuration and installation requires
highly skilled personnel adept at configuration of
many network elements. Where interactions
between network nodes (switches, routers, etc.)
are complex, a more systems-based approach
encompassing elements of simulation is required.
With the current programming interfaces on
much of today’s networking equipment, this is
difficult to achieve.
In addition, operational costs involved in provisioning
and managing large multivendor networks
covering multiple technologies have been
increasing over recent years, while the predominant
trend in revenue for operations has been
decreasing. Coupled with increasing scarcity of human resources and increasing costs of real
estate, this “perfect storm” for service providers
is leading to renewed interest in solutions that
can unify network management and provisioning
across multiple domains. A new network model
is required to support this.
The term software-defined networking (SDN)
has been coined in recent years. However, the
concept behind SDN has been evolving since
1996, driven by the desire to provide user-controlled
management of forwarding in network
nodes. Implementations by research and industry
groups include Ipsilon (proposed General Switch
Management protocol, 1996), The Tempest (a
framework for safe, resource-assured, programmable
networks, 1998) and Internet Engineering
Task Force (IETF) Forwarding and
Control Element Separation, 2000, and Path
Computation Element, 2004. Most recently,
Ethane (2007) and OpenFlow (2008) have
brought the implementation of SDN closer to
reality. Ethane is a security management architecture
combining simple flow-based switches
with a central controller managing admittance
and routing of flows. OpenFlow enables entries
in the Flow Table to be defined by a server
external to the switch. SDN is not, however, limited
to any one of these implementations, but is
a general term for the platform.
For clarity, SDN is described in this article
with the Open Networking Foundation (ONF)
[1] definition: “In the SDN architecture, the control
and data planes are decoupled, network intelligence
and state are logically centralized, and the
underlying network infrastructure is abstracted
from the applications.”
SDN focuses on four key features:
• Separation of the control plane from the
data plane
• A centralized controller and view of the
network
• Open interfaces between the devices in the
control plane (controllers) and those in the
data plane
BACKGROUND: WHY SDN?
The fundamental purpose of the communication
network is to transfer information from one
point to another. Within the network the data
travels across multiple nodes, and efficient and
effective data transfer (forwarding) is supported
by the control provided by network applications/services.
NETWORKING THE OLD WAY
In traditional networks, as shown in Fig. 2, the
control and data planes are combined in a network
node.
The control plane is responsible for configuration
of the node and programming the paths
to be used for data flows. Once these paths have
been determined, they are pushed down to the
data plane. Data forwarding at the hardware
level is based on this control information.
In this traditional approach, once the flow
management (forwarding policy) has been defined,
the only way to make an adjustment to the policy
is via changes to the configuration of the devices.
This has proven restrictive for network operators
who are keen to scale their networks in response
to changing traffic demands, increasing use of
mobile devices, and the impact of “big data.”
CONCLUSION
SDN has emerged as a means to improve programmability
within the network to support the
dynamic nature of future network functions. As
bandwidth demand escalates, the provision of
additional capabilities and processing power with
support for multiple 100GE channels will be
seamless through an SDN-based update and/or
upgrade. SDN promises flexibility, centralized
control, and open interfaces between nodes,
enabling an efficient, adaptive network.
In order to achieve this goal, a number of
outstanding challenges must be resolved. In this
article we have presented a discussion of a number
of challenges in the area of performance,
scalability, security, and interoperability. Existing
research and industry solutions could resolve
some of these problems, and a number of working
groups are also discussing potential solutions. In addition to these, the hybrid programmable
architecture could be a means to
counter performance and scalability issues introduced
by SDN. The objective of the model is to
optimize flow processing in the network.
The original data networks were formed out
of a combination of computing devices with data
and network nodes to transfer this data between
the source and destination. The ability to provide
“X”-as-a-service (XaaS) through virtualization
technology has increased the volume of data
in the network. This has set a baseline for a new
communication method by pushing computation
into the network devices, increasing machine-tomachine
communications.
The future of networks will be shaped around
this progression. The goal is to provide effective
communications and services where network,
data, and computation are fused into a service
architecture. In the future, for a specific process,
data will request the computing, storage, and
connection it requires before launching the
application. The location of the network elements
might be distributed physically and virtually,
but this will be entirely opaque to the end
user. All the user will observe is the quality of
delivery of the requested service.
SDN will contribute to this vision of future
communications. However, significant issues
must be addressed in order to meet expectations.
Indeed, consideration of the potential for
application-driven networks might lead us to
wonder whether SDN as currently envisioned is
even sufficient. Nevertheless, it is certain that
SDN is here to stay as an evolutionary step,
paving the way toward a highly optimized ubiquitous
service architecture.