TRANSACTIONS ON EMERGING TELECOMMUNICATIONS TECHNOLOGIES
Trans. Emerging Tel. Tech. (2016)
Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/ett.3084

SPECIAL ISSUE ARTICLE

DevOps for network function virtualisation: an
architectural approach
Holger Karl1 , Sevil Dräxler1 , Manuel Peuster1 , Alex Galis2 , Michael Bredel3 ,
Aurora Ramos4*, Josep Martrat4 , Muhammad Shuaib Siddiqui5 ,
Steven van Rossem6 , Wouter Tavernier6 and George Xilouris7
1 Paderborn University, Paderborn, Germany
2 University College London, London, UK
3 NEC Laboratories Europe, Heidelberg, Germany

4 ATOS Spain, Madrid, Spain
5 Fundació i2CAT, Barcelona, Spain
6 Ghent University – iMinds, INTEC, Gehnt, Belgium
7 Institute of Informatics and Telecommunications, NCSR ‘Demokritos’, Greece

ABSTRACT
The Service Programming and Orchestration for Virtualised Software Networks (SONATA) project targets both the ﬂexible
programmability of software networks and the optimisation of their deployments by means of integrating Development
and Operations in order to accelerate industry adoption of software networks and reduce time-to-market for networked
services. SONATA supports network function chaining and orchestration, making service platforms modular and easier to
customise to the needs of different service providers, and introduces a specialised Development and Operations model for
supporting developers. Copyright © 2016 John Wiley & Sons, Ltd.
*Correspondence
Aurora Ramos, ATOS Spain, Madrid, Spain.
E-mail: aurora.ramos@atos.net
Received 6 April 2016; Revised 14 June 2016; Accepted 27 June 2016

1. INTRODUCTION
In the ongoing research discussion about the structure
and architecture of 5G networks, a basic trend is evident: the network will have to be much more ﬂexible
and open to profound changes of its operation to match
evolving demands. These demands will not just evolve
quantitatively but also qualitatively in that trafﬁc with different, yet unforeseeable, characteristics will have to be
supported. An ideal approach to increase ﬂexibility of a
system is to scale software, more precisely, of reconﬁgurable and replaceable software. This is the core notion
of providing conventionally ﬁxed network functions via
software, typically in a virtualised environment: network
function virtualisation (NFV) [1]. It is a transformation that
relies on software-based innovation on top of more generic
telecommunications hardware.
This disruptive architecture is challenging to put into
practice. The value chain of the provisioning and operation of networks will change and needs to be supported
by a technical environment. This environment will need to
Copyright © 2016 John Wiley & Sons, Ltd.

support a decrease in development time and improved network operations, including further overlapping functions.
Opportunity also arises as the role of providing network
functionality taken up by a broader range of actors, no
longer limited to the vendor of network equipment. This
allows a much higher degree of customisation of network
functions for the service developer, empowering them with
a better deployment and greater control of their service.
Conversely, the role of the operator of a network changes
as well: instead of just running a system with essentially
ﬁxed functionality where only parameterisation is necessary, the dynamics provisioning and deploying new functions become daunting. If, in particular, provisioning will
become a frequent action rather than a unique event, it must
be supported by automation that spans across the development and the operational cycle of network functions, as
well as orchestration of needed resources—the technical
role of an ‘orchestration platform’ becomes important and
has received considerable attention.
Taking such a comprehensive view of development and
operation of complex software is indeed a common trend of

H. Karl et al.

modern software engineering; it is commonly referred to as
‘DevOps’ (a contraction of Development and Operations,
to highlight the close integration of these two processes).
It has not, however, taken root in the actual practice of the
telecommunication industry with respect to the operation
of networks in an open fashion. This is the core shortcoming that the SONATA project [2] addresses, providing an
integrated development and operational process for virtualised network functions, along with the necessary supporting tools like a service-adaptable and resource-adaptable
orchestration platforms.
Section 2 provides a perspective of previous results
related to SONATA (including research work, standardisation initiatives and common trends in an initial
generation pre-production open-source and commercial
architectures). Section 3.1 looks into how SONATA’s challenges would have to be reﬂected in a broader 5G architectural context. Speciﬁc requirements and challenges arising
from various use cases are presented in Section 4. Finally,
Section 5 describes the actual SONATA contribution to
a ‘softwarised’ network architecture. Section 6 concludes
and previews future activities.

2. RELATED WORK
Network function virtualisation [3] is currently on operators’ road maps for deployment, but at the time of
writing mostly in a proof-of-concept stage of adoption.
However, the arrival of software networks, supported by
software-deﬁned networking [4] and NFV technologies,
is a universally agreed milestone in the evolution of
telecommunications. Several open-source, standardisation
and commercial solutions are being developed for NFV
management and orchestration [5] [‘MANO’, as European
Telecommunications Standards Institute (ETSI) terms it] in
anticipation of its widespread operational use.
2.1. Software development processes and
tools for network function virtualisation
There is surprisingly little material available of software development for network functions or resulting services, let alone concrete tools to this end. The Unifying
Cloud and Carrier Networks (UNIFY) project [6] delivers some ﬁrst insights on how to apply the DevOps
model to NFV. They provide a multicomponent debugging tool called Epoxide [7]. Compared with these tools,
SONATA’s approach is a step forward and combines a
powerful software development kit (SDK) with a ﬂexible service platform to provide end-to-end support for
service developers.
2.2. Orchestration platforms
Simple cloud managers like provide the basic functionality
to deploy and manage single predeﬁned virtual machines
but cannot handle composed services. Other solutions,

like OpenStack Heat [8], Terraform [9] and Application
Deployment Toolkit (ADT) [10], are able to deploy entire
services but do not focus on network function-speciﬁc
needs. Projects directly focusing on NFV service orchestration can be divided into three categories. The ﬁrst
consists of research projects, like Network Functions as
a Service over Virtualised Infrastructures (T-NOVA) [11]
and UNIFY [6]. The second category consists of opensource projects such as OpenMANO [12] or OpenStack
Tacker [13] or open software-deﬁned infrastructure [14].
A third category is an initial generation of commercial solutions put forth by telecommunications vendors
(expanding their platform, software and integration services to adapt to the disruptive change in the value chain)
and supporting technology partners (classically in IT,
cloud, etc. but recognising new market opportunities as
telecom infrastructure becomes virtualised and softwarebased). Regardless of their origin, they offer similar functionality and characteristics when conﬁned to the ETSI
NFVO speciﬁcation, and signiﬁcant differentiation is more
apparent when comparing their larger ecosystems that
extend beyond the scope of orchestration. This is consistent with the business strategy to provide all NFV-related
needs (through their own portfolio or partnerships).
UNIFY’s architecture [6] aims at automated and
dynamic service creation and recursive resource orchestration. Its global orchestrator includes optimisation
algorithms for placement of service components; servicespeciﬁc actions related to placement and scaling are
deployed as a service component. T-NOVA is capable of
managing services distributed across several data centres
but only provides a simple, rule-based system to control
the scaling and placement of deployed services. It also provides a marketplace in which predeﬁned services and virtual network functions (VNFs) can be traded. OpenMANO
and OpenStack Tacker aim to be reference implementations of the MANO layer deﬁned in the ETSI NFV Industry
Speciﬁcation Group (ISG) architecture [15], but both are
at the beginning of their development. Other orchestration
solutions are Cloud4NFV [16] and vConductor [17].
All presented orchestration tools, to a great extent,
follow the same principle and try to build a single orchestration solution for different types of services. This creates
multiple restrictions for service developers as they cannot
inﬂuence the orchestration process as such. Some of the
existing platforms allow expressing a limited and predeﬁned set of preferences regarding monitoring, scaling
and so on. within function descriptions. However, actively
inﬂuencing service-speciﬁc decisions, for example,
placement and scaling of services and their components,
is not fully supported by any of them. This is possible
with SONATA’s service platform using function and
service-speciﬁc manager programmes deﬁned by the
service developer.
Trans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

H. Karl et al.

3. SONATA SCOPE AND AN
OVERALL 5G ARCHITECTURE

3.2. 5G multi-service management

3.1. Scope of SONATA functionality
SONATA aims at increasing the ﬂexibility and programmability of 5G networks with a novel Service Development
Kit and a novel modular Service Platform and Service
Orchestrator; it will bridge the gap between telecom business needs and operational management systems.
The scope of SONATA is depicted in Figure 1 where
the outcome of the project is represented by the Service Development Kit, the Management System and
the Service Platform including a customisable Service
Orchestrator, a Resource Orchestrator, a Service Information Base along with various enablers; we use here
the ETSI reference model as a terminological framework. In fact, while ETSI’s division into Service and
Resource Orchestrator can be mapped onto SONATA’s
service platform, SONATA is much more ﬂexible in this
regard and allows not only replacing these orchestration aspects individually, but even to change the division
of work between these functions if so desired. This is
achieved by SONATA’s microservices-based architecture
(Section 5.1.1); for example, the resource orchestrator corresponds by and large to SONATA’s default placement
plug-in.

SONATA advocates a consistent view of 5G network
and compute functions, encompassing a wide conceptual
range of such functionality. SONATA functionality covers the multi-service control layer and partially integrated
management and operation layer and the application and
business services layer. SONATA is also capable of incorporating widely heterogeneous physical resources: various
access networks (esp., radio), aggregation and core networks, software networks, data centre networks and mobile
edge computing clouds.
A multi-service control layer is responsible for the
creation, operation and control of multiple dedicated communication network services running on top of a common
infrastructure. SONATA’s functionality [18] for this layer
includes infrastructure abstraction, infrastructure capability discovery, catalogues and repositories, large number
of service and resource orchestration functions as plugins, information management functionality and enablers
for automatic reconﬁguration of running services, that is,
part of the integrated management layer.
The business function layer maintains 5G applicationrelated functions, organised in repositories, and DevOps
tools necessary for the creation and deployment of services. SONATA’s functionality for this layer includes
DevOps functionality: catalogues, monitoring data
analysis tools, testing tools, packaging tools, editors

Figure 1. SONATA scope.
Trans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

H. Karl et al.

Figure 2. SONATA’s relationship to heterogeneous underlying infrastructures.

and basic functionality for application and service
programmability.
Figure 2 depicts the way in which SONATA manages
various underlying systems. The core idea is to endow
SONATA with several infrastructure abstractions, each of
which is custom-tailored to the particular needs of the
underlying infrastructure. This allows both simple and
complex of Virtual Infrastructure Managers (VIMs) to be
used, as well as entire networks or single, for example,
OpenStack instances to be integrated.
In summary, SONATA’s main contribution to 5G networking is efﬁcient integration of service programmability,
domain orchestration functionality and DevOps functionality. This will maximise the predictability, efﬁciency,
security and maintainability of development and operational processes around virtualised network functions and
chain services.

4. USE-CASE-DRIVEN CHALLENGES
AND REQUIREMENTS
4.1. Use cases and resulting requirements
The use cases facilitate the identiﬁcation of requirements
of the SONATA framework while highlighting its full

potential for future software networks. SONATA use cases
[19] encompass a wide range of Information and Communication Technologies (ICT) domains and their network
services as follows:
Internet of Things: demonstrates SONATA’s ability
to monitor, classify and optimise Internet of Things
network trafﬁc as an enabler of Smart City ecosystem.
Virtual Content Delivery Networks (CDN): manifests
SONATA’s capabilities to enhance a virtual CDN
service with elasticity and programmability.
Industrial networks: providing guaranteed, resilient
and secure service delivery in vertical industrial networks, such as a wind park.
Virtual Evolved Packet Core: exhibits and assesses
the SONATA’s competence to enhance a virtual
Evolved Packet Core service in an long-term evolution mobile network.
Personal security service: targets the system’s potential to offer on-demand network security services to
multiple tenants/users with differentiated policies and
enable secure network access.
Separate client and hosting service providers:
demonstrates SONATA’s support for internetworking between a client service provider and a hosting
service provider to accomplish an end-to-end service.
Trans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

H. Karl et al.

Based on these use cases, we can distil high-level
DevOps requirements for both the overall process of developing network functions (SDK) as well as the runtime
platform steering the management and life cycle of such
functions in an actual infrastructure/network—commonly
called the orchestrator.
(1) Development model requirements
For the development model, the following requirements are elicited:
Support for arbitrary network functions: Support for all types of network functions must be
possible, irrespective of the technical domain.
For example, both core networks as well as
wireless fronthaul networks must be supported.
To express the different requirements, suitable annotation for network function and services must exist. Scenarios that pertain to
both network functions in the narrow sense
as well as generic (mobile) edge/distributed
cloud computing are in the purview of the
SONATA system.
Network functions form data ﬂow graphs: Individual functions will not sufﬁce to address all
needs; it is needed to group functions into
general graphs of functions which then constitute the actual service (i.e network service).
This is in-line with current NFV speciﬁcations
(Internet Engineering Task Force (IETF) service
chains [20]; ETSI service graphs).
Reuse by recursive deﬁnitions: Reusing such
service graphs mandates a recursive notion—
a new service can be formed not only by
using primitive network functions but also
by reusing other, simpler services. Hence, a
recursive approach has to prevail in the software development concepts and description
techniques.
Reuse by dynamic composition: Recursive deﬁnitions are the ﬁrst stepping stone. This concept
becomes really powerful if existing network
functions or service can be dynamically reused
and combined into new services without requiring new deployments (assuming the constituting
network functions is multi-user-capable).
Policy-driven support for developer’s service:
Each service will not only have its own code
that deals with the actual data ﬂows but also
have additional code that takes care of the
deployment and execution process in a manner tailored to the speciﬁc requirements of the
service—we refer to such code here simply as
‘service-speciﬁc code’ and make this notion
more concrete in later sections. For example,
scaling or placing a complex service requires
decision logic that is rather speciﬁc to the
particular service; similarly, life cycle manageTrans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

ment is highly service-speciﬁc (e.g. mandating a
particular order in which virtual machines comprising a service have to be booted and initialised). This is impossible to achieve in a
one-size-ﬁts-all fashion.
As a consequence, we need to support both aspects of
a service’s code: the actual data-oriented code as well
as the caretaking code. This materialises as supporting tools for the developer, in speciﬁcation for what
to ship as a service, and how to be rendered by the
orchestration platform, which must be able to execute such service-speciﬁc code in a proper context
and with proper data.
(2) Orchestration platform requirements
Similarly, for the orchestration platform, we can
identify the following core requirements:
No one solution ﬁts all: The large variability of services to be orchestrated, along with
their own individual caretaking functions, mandates that the orchestration platform must be
able to execute service-speciﬁc code, but must
ensure proper information hiding and isolation between these code pieces from different
services.
Design to support for evolution and maintainability: Assuming that an orchestration platform will ever have a stable, ﬁnal form seems
overly optimistic. Rather, the platform itself
must be able to be extended by additional functionality or to replace existing one. This precludes a monolithic design of the platform itself.
A microservices-based approach, with individual modules executing particular aspects
(together with the individual functions caretaking code) of an orchestration cycle, seems more
promising for adoption by network operators
and their bespoke conﬁguration of the platform.
Support real-world’s manifold systems: An
orchestration platform should be able to support a wide range of actual infrastructures on
which services will be deployed. This ranges
from bare-metal over simple hypervisors to fullﬂedged OpenStack installations conceived of
a single logical node. A particular example
is sliced infrastructures, where the underlying
infrastructure is divided into isolated ‘slices’,
each of which has its own guaranteed resources
[21]. In the simplest case, SONATA obtains
new slices from an external slicing system (in
the extreme case, one slice per service is conceivable). Alternatively, it might be promising
to integrate slicing functionality directly into
an orchestrator.
Division and recursion on platform layer: Akin
to the idea of recursion to describe services,
it is promising to conceive infrastructure and
orchestration platform itself in a recursive

H. Karl et al.

way, as well. Sub-orchestrators can then be
in charge of different domains or different
operator networks, working together under the
auspice of a higher-level orchestrator. The challenge is to have the orchestrator itself expose
an infrastructure-like interface. This recursive
approach should also combine nicely with a
slicing system.
These high-level requirements and associated challenges are addressed by SONATA—the following sections
show how.

5. SOFTWARE-NETWORK
TECHNOLOGIES
SONATA project’s main goal is to increase the ﬂexibility and programmability of 5G networks in order to
bridge the gap between telecom business needs and operational management systems. In this chapter, an overview
of the SONATA architecture is provided, including a short
description of the main components. The service programming and orchestration framework consist of the SDK, the
service platform and different catalogues storing artefacts
that can be produced, used and managed by the SONATA
system. Services developed and deployed by this system
run on top of the underlying infrastructure accessible to
the SONATA system via a wide range VIMs; SONATA is
fairly agnostic to the particular VIMs.
SONATA’s system design is based on the DevOps workﬂow, which is supported by the integration between the
SDK and the service platform. This workﬂow implies
continuous deployment and continuous integration during
service development. The main entity exchanged between
the SDK and the service platform is the service package
to be deployed and runtime information like monitoring
data and performance measurements regarding the service
package, which is provided to the service developer during the development phase, as well as the runtime. This
information can be used for optimising, modifying and
debugging the operation and functionality of services.
The main characteristics of the SONATA architecture
are explained in Section 5.1, followed by covering the four
main actor roles in the SONATA platform (Section 5.2).
For a full description of the current state of the SONATA
architecture, please refer to the corresponding deliverable
[18] at the time of this writing, to be updated during the
course of the project.
5.1. Some architectural aspects
(1) A microservices-based orchestration platform
The high-level service deployment procedure is
illustrated in the service platform. Each VIM/WAN
Infrastructure Manager (WIM) provides the controlling service platform a view of the available
resources and capabilities of its underlying infrastructure/network. A gatekeeper module in the service
platform is responsible for processing the incoming

requests. The service platform receives the service
packages implemented and created with the help
of SONATA’s SDK and is responsible for placing,
deploying, provisioning, scaling and managing the
services on existing cloud infrastructures. For this
purpose, it has modules for orchestrating and managing the complete service chain, as well as managing
on the VNF level. All artefacts needed to deploy the
service can be fetched from catalogues and repositories. The platform can also provide direct feedback
about the deployed services to the SDK, for example,
monitoring data about a service or its components.
SONATA’s service platform is designed with full
customisation possibility, providing ﬂexibility and
control to both operators and developers. The core
mechanism for this is a microservices-based plug-in
architecture (Figure 3): all functionality that is to be
provided by an orchestrator is assigned to speciﬁc
plug-ins, all of which are connected to a message
bus that ensures correct delivery semantics of all
control messages between these plug-ins. Some of
these plug-ins implement exactly one function per
orchestrator (e.g. the conﬂict resolver plug-in, which
ensures that any possible resource conﬂicts between
services are resolved in a consistent, service-neutral
fashion).
Other plug-ins can be customised by the deployed
service itself with caretaking code: these plug-ins
then act as an executive (akin to Microsoft Window’s
operating system concept) for this service-speciﬁc
code.
The service developer can ship the service package
to the service platform together with service-speciﬁc
or function-speciﬁc caretaking code, expressing and
realising requirements and preferences. Such caretaking code is referred to in SONATA as servicespeciﬁc managers and function-speciﬁc managers,
respectively. Service-speciﬁc managers and functionspeciﬁc managers can inﬂuence the service and VNF
life cycle management operations, for example, by
specifying desired placement or scaling behaviour.
This grants the developer increased ﬂexibility, control and resilience of their service.
By virtue of a modular design in the Management
and Orchestration Framework of the service platform, the service platform operator can customise it,
for example, by replacing the conﬂict resolution or
information management modules. SONATA’s service platform is described in more detail in project
deliverable D2.2 [18]. This could be an operator
replacing modules of the conﬁgurable service platform with alternatives to ﬁt their software network
management requirements and preferences.
(2) Recursive orchestration
A recursive structure can be deﬁned as a design,
rule or procedure that is (partially) explained using
a simpliﬁed version of itself. In a network service context, this recursive structure can either be a
Trans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

H. Karl et al.

Figure 3. Main plug-ins into SONATA’s orchestrator.

Figure 4. Service deployment using the SONATA framework.

speciﬁc part of a network service or a repeated part
of the deployment platform. Although different challenges can be thought of, the general idea of reusing
Trans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

existing patterns could reduce complexity and even
add more ﬂexible possibilities for extending the
service [22]. In Figure 4, recursive orchestration is

H. Karl et al.

shown as a SONATA service platform delegating
(part of) the requested service to another instance of
a SONATA platform using a dedicated infrastructure
adaptor.
Reclusiveness also leads to an easier management
of scalability. Monolithic software entities are prone
to performance limitations from a certain workload
onwards. Scaling by delegating parts of the service
to multiple instances of the same software block is
a natural way to handle more complex and larger
workloads or service graphs. If this reclusiveness is
taken into account from the beginning of the development, the advantages of this approach will come at
a minimal cost.
(3) An software development kit for network function
virtualisation
The SDK supports service developers by providing a service programming model and a development tool-chain. Figure 5 shows an overview of the
foreseen SDK components. SONATA’s SDK design
allows developers to deﬁne and test complex services
consisting of multiple network functions, with tools
that facilitate custom implementations of individual
network functions. The implemented artefacts are
stored in the developer’s private catalogues. Moreover, service components can easily be obtained from
external catalogues using the foreseen interfaces. The
obtained artefacts can be directly used in a service
or after being modiﬁed and tested using the SDK
development tools. The service components and all
the information necessary for deployment and execution of a service are bundled together into a package.
The service package can be handed over to the service platform for actual deployment and for testing,
debugging and proﬁling purposes.
5.2. The main actors in SONATA
(1) End users
It is the entity that consumes the network service. Thus, an end user can be a private person or

Figure 5. Main components of SONATAs SDK.

an enterprise, or even a network service provider
or content provider. In the SONATA ecosystem,
the end user requests the required network services
from the network service developer—Figure 6 only
shows the simpler case. That is, the end user and
the network service developer establish a customer–
provider relationship that is regulated by a service
level agreement.
(2) Service developer
It is the developer of a software artefact, for
example, a VNF or, more speciﬁcally, a network service, which can be composed of one or more VNFs.
The SONATA SDK shall enable a DevOps environment for the development of network services.
The network service developer can use the editor,
debugging and packaging tools in the SONATA SDK
along with VNF catalogues to compose a network
service that can then be moved to the SONATA service platform for deployment and execution. That
is, the network service developer interacts with the
end user for providing the network services and also
interacts with the SONATA service platform operator
for the deployment and operation of those network
services. However, in the value chain, the developer
and provider of such a service can of course be
separate entities.
(3) Service platform operator
The service platform operator runs the SONATA
platform that manages the execution of network services. The SONATA service platform receives a network service in form of a package that is validated
through a series of quality assurance tests prior to
their storage in the service platform catalogue. In
addition to validation, the service platform operator also manages the deployment and execution of
network services on virtual resources made available by an infrastructure operator. As the SONATA
service platform supports reclusiveness, an operator can manage multiple SONATA service platform
instances, as well. Hence, a responsibility of maintaining smooth operations for a network service, with
respect to its corresponding service level agreement,
lies on the service platform operator. On one side, the
service platform operator interacts with the SONATA
SDK for network services reception and, on the other
side, interacts with the infrastructure operator for
their deployment and execution.
(4) Infrastructure operator
It is the entity that actually operates the physical infrastructure, including computational, communication and storage resources. The SONATA
ecosystem does not distinguish between infrastructure owner and infrastructure operator and treat
them as the same. This is because the main focus
of SONATA is to reduce time-to-market for network services by accelerating and facilitating service
development, management and orchestration with
DevOps adoption, all while ensuring features such as
Trans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

H. Karl et al.

Figure 6. Main components and roles.

multi-tenancy, reclusiveness and virtual network
slices are supported. The infrastructure operator
interacts heavily with the service platform operator
as the network services are actually executed in the
physical infrastructure. This interaction is enabled by
the infrastructure abstraction layer in the SONATA
service platform. It is worth mentioning that in most
scenarios, the infrastructure operator will be the same
as the service platform operator. However, SONATA
also supports a use case where the system is engaged
by separate entities for these roles, supporting more
complex telecom provider business models.

6. CONCLUSIONS AND
FUTURE WORK
The SONATA architecture and resulting system push innovation in the evolving state of the art of NFV MANO and
network service development. From the approach outlined
earlier, we can draw four main conclusions.
Modular and customisable MANO plug-in architecture: the architecture structure provides hitherto unseen
levels of ﬂexibility. Third-party developers are empowered
with control over speciﬁc orchestration and management
functionalities pertaining to their own service without
being able to inﬂuence other services. Similarly, a network
operator has a wide range of control over how the service
platform should behave. For example, the service platform
effortlessly encompasses ETSI’s separation into resource
and service orchestration, but other orchestration models
are supported as well. This boils down into a new notion
of Orchestration-as-a-Service (OaaS): the orchestration
process itself can be customised on demand by the provider
of a service but stays under the operator’s control.
Interoperable, agnostic framework: the resulting
framework (incorporating both service platform and
Trans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

developer support) is agnostic along multiple axes: it supports multiple VIMs, multiple vendors, deployment at
multiple sites. It is also agnostic to whether these functions
are used in different kind of networks (mobile, optical, etc.)
or in the fronthaul or backhaul part. It is agnostic with
respect to whether these services are stateless or stateful.
Efﬁcient network service development and DevOps:
SONATA provides service developers with an SDK for
efﬁcient creation, deployment and management of network
services composed of VNFs along with the required chaining on the platform. The integration of SDK and service
platform links an existing gap between service developers
and operators.
5G integration: SONATA easily embraces current 5G
architecture developments and plays as a good citizen
in the entire ecosystem. For example, slicing is well
supported in various fashions, interacting with external
slicing systems as is bespoke network conﬁguration for
industry verticals. Recursion support allows stacked tenant and wholesale deployments in new software networks
business models.
DevOps prototype: SONATA is an ongoing research
project with its ﬁrst code release scheduled early July
2016. The consortium plans to demonstrate a ﬁrst prototype showcasing the complete life cycle of a network
service in the NFV environment with continuous monitoring by end of year 2016 as well as by exercising several
pilot applications that will be chosen among the use cases
described in Section 4. The ﬁnal prototype with full feature
set is scheduled for end of year 2017. Using this prototype with its continuously increasing capabilities, we will
elaborate on the reduction of development time. Using the
SONATA systems, we expect the time to deploy a new
network service to be much faster compare with common approaches today. Moreover, we expect a signiﬁcant
reduction of code to write in order to operate network

H. Karl et al.

functions and services compared with existing technologies. The proposed DevOps approach raises more technical
challenges, for example, related to the update process of
running network functions and service. To this end, we
will investigate different approaches—again looking at the
complete life cycle—and evaluate them in terms of performance, service continuity, resiliency and fault tolerance.
Finally, we will compare the SONATA system with existing solutions with respect to resource efﬁciency and overall
service performance.

REFERENCES
1. Chiosi M, Clarke D, Willis P, Reid A, et al. Network
functions virtualisation. Technical Report, White paper
at the SDN and OpenFlow World Congress, ETSI, 2012.
2. SONATA project, 2015. Available from: http://
sonata-nfv.eu/ [accessed on April 2016].
3. Mijumbi R, Serrat J, Gorricho J, Bouten N, De Turck
F, Boutaba R. Network function virtualization: stateof-the-art research challenges. IEEE Communications
Surveys & Tutorials 2016; 18(1): 236–262.
4. McKeown N, Anderson T, Balakrishnan H, Parulkar G,
Peterson L, Rexford J, et al. Openﬂow: enabling innovation in campus networks. ACM SIGCOMM Computer
Communication Review 2008; 38(2): 69–74.
5. ETSI. NFV MANO. Available from: http://portal.etsi.
org/portal/server.pt/community/NFV/367?tbId$=$79
[accessed on April 2016].
6. Unify project, 2015. Available from: https://www.
fp7-unify.eu/ [accessed on April 2016].
7. Lévai T, Pelle I, Németh F, Gulyás A. EPOXIDE: a modular prototype for SDN troubleshooting. In ACM Conference on Special Interest Group on Data Communication,
London, 2015.
8. OpenStack heat, 2015. Available from: https://wiki.
openstack.org/wiki/Heat [accessed on April 2016].
9. Terraform, 2015. Available from: https://www.terraform.
io [accessed on April 2016].
10. Keller M, Peuster M, Robbert C, Karl H. A topologyaware adaptive deployment framework for elastic applications. In 17th International Conference on Intelligence
in Next Generation Networks (ICIN), Venice, Italy, 2013.

11. Xilouris G, Trouva E, Lobillo F, Soares JM, Carapinha J,
McGrath M, et al. TNOVA: a marketplace for virtualized
network functions. In European Conference on Networks
and Communications (EuCNC), Bologna, Italy, 2014.
12. OpenMANO, 2015. Available from: https://github.com/
nfvlabs/openmano [accessed on April 2016].
13. OpenStack tacker, 2015. Available from: https://wiki.
openstack.org/wiki/Tacker [accessed on April 2016].
14. Mamatas L, Clayman S, Galis A. A service-aware virtualized software-deﬁned infrastructure. IEEE Communications Magazine 2015; 53(4): 166–174.
15. ETSI NFV ISG. GS NFV-MAN 001 V1.1.1 network
function virtualisation (NFV); management and orchestration, Dec. 2014.
16. Soares J, Dias M, Carapinha J, Parreira B, Sargento
S. Cloud4NFV: a platform for virtual network functions. In IEEE 3rd International Conference on Cloud
Networking (CloudNet), Luxemburg, 2014.
17. Shen W, Yoshida M, Minato K, Imajuku W. Conductor: an enabler for achieving virtual network integration
as a service. Communications Magazine 2015; 53 (2):
116–124.
18. Architecture design: deliverable D2.2—SONATA project. Available from: http://sonata-nfv.eu/sites/default/
ﬁles/sonata/public/content-ﬁles/pages/SONATAD2.
2ArchitectureandDesign.pdf [accessed on April 2016].
19. Use cases and requirements—deliverable 2.1—SOTATA
project. Available from: http://sonata-nfv.eu/use-cases
[accessed on April 2016].
20. IETF service chaining. Available from: https://
datatracker.ietf.org/wg/sfc/documents/ [accessed on
April 2016].
21. Argyropoulos C, Mastorakis S, Giotis K, Androulidakis
G, Kalogeras D, Maglaris V. Control-plane slicing methods in multi-tenant software deﬁned networks. In 2015
IFIP/IEEE International Symposium on Integrated Network Management (IM), Ottawa, 2015; 3–4.
22. Szabo R, Kind M, Westphal FJ, Woesner H, Jocha D,
Csaszar A. Elastic network functions: opportunities and
challenges. IEEE Network 2015; 29(3): 15–21.

Trans. Emerging Tel. Tech. (2016) © 2016 John Wiley & Sons, Ltd.
DOI: 10.1002/ett

View publication stats