FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
Enabling Information Interoperability
in the Future Internet
Martín Serrano, Mícheál Ó Foghlú and William Donnelly
Waterford Institute of Technology - WIT
Telecommunictions Software and Systems Group - TSSG
Co. Waterford, Ireland
{jmserrano, mofoghlu, wdonnelly}@tssg.org
Abstract. The Future Internet is currently seen as an opportunity to improve the
network infrastructure addressing service-oriented, social trends and economic
commitments. Future Internet and next generation communications systems
challenges mainly demand end user requirements, personalized provisioning,
service-oriented performance, service-awareness networking, information
interoperability and data models integration. This paper focuses on information
interoperability and cross domain information managing Internet systems. Research
activity and results about semantic enrichment tasks for management information
contained in both enterprise and networking information and data models
respectively is described. Ontologies are used to support reusable, common and
manageable service and network information for service composition and
management operations in the inference plane in Future internet systems. An
introductory application scenario on current service agnostic Internet is depicted.
Keywords. Interoperability, Cross-domain, Future Internet, Knowledge Engineering,
Semantics, Ontologies, Networks and Services, Management.
1. Introduction
Convergence towards Internet technologies for communications networks and
services has been a clear trend in the Information and Communications Technology
(ICT) domain in the past few years. Although widely discussed, this trend has not fully
run its course in terms of implementation, due to many complex issues involving
deployment as result of non-interoperable aspects where social, economic and political
dimensions take place, all these issues a matter of end-user demands and requirements.
The Future Internet (FI) is currently seen as an opportunity to improve the network
infrastructure addressing service-oriented, social trends and economic commitments.
So challenges in the future communications systems mainly Internet-based systems
demand, in terms of end user requirements, personalized provisioning, service-oriented
performance, and service-awareness networking. To support these demands
information interoperability and data model integration are crucial.
A visionary perspective for what the Future Internet looks like has been described in
previous works [1][2][3]. The intention in this paper is not to define what the Future
Internet is, but rather to view the Future Internet in a service-oriented manner, coming
through a revision about the role knowledge engineering can play to satisfy part of the
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
mentioned challenges. In Future Internet, services and networks follow a common
guideline; provide solutions in form of implemented interoperable mechanisms.
Communications networks have undergone a radical shift from a traditional circuit-
switched environment with heavy/complex signalling focused on applications-oriented
perspective, towards a converged service-oriented architecture space (SOA), mostly
Internet interaction by customer as end-user and network operators as service
providers. The business benefits of this shift significantly reflects cost reduction and
increase systems flexibility to react to user demands more efficiently and by replacing,
in a best practice case, a plethora of proprietary hardware and software platforms with
generic solutions supporting standardised development and deployment stacks. As an
example of this shift, the emergence and wide-scale deployment of wireless access
network technologies calls into question about the viability of basing the future Internet
on IP and TCP – protocols that were never intended for use across highly unreliable
and volatile wireless interfaces (information and data exchange).
Research initiatives addressing this SOA requirement argue that the future lies in
layers of overlay networks that can meet various requirements whilst keeping a very
simplistic, almost unmanaged, IP for the underlying Internet, GENI NSF-funded
initiative to rebuild the Internet [4]. Others argue that the importance of wireless access
networks requires a more fundamental redesign of the core Internet Protocols
themselves [5][6]. Whilst this debate races nothing is a clear outcome in terms of
information interoperability or data models sharing.
We follow the idea that service agnostic design are not anymore a way to achieve
interactive solutions in terms of service composition and information sharing
capabilities for heterogeneous infrastructure support. A narrow focus on designing
optimal networking protocols in isolation is too limited. Instead, a more holistic and
long-term view is required, in which networking issues are addressed in a manner
various protocols delivering communications services can be supported, meeting
rapidly changing communities of users needs. This new holistic view increasingly stops
to become a matter of critical infrastructure. Network operators are today coming to
realise lack in the promise of simpler all-IP networks, where new integrated Internet
services are easier and quicker to design, develop, deploy and manage.
Linked-Data Layer
Service-Oriented
Archtecture
Heterogeneous
Infrastructure
Figure 1. Knowledge-Based Service Composition View – Information Interoperability.
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
The figure 1 depicts the mentioned service-aware Future Internet holistic view and
its implementation relies on the inference plane [7], a plane where the exchange of
information facilitates knowledge-driven support and generation of composed services
with operations by enabling interoperable management information. So the move
towards converged IP-based communications networks increases providing solutions to
a number of significant technical issues by using more standard information exchange
promoting information interoperability and allowing the networks can be managed
effectively and most important offering open opportunities for a user knowledge-based
service-oriented support can have a fundamental impact on future communications
networks and services.
This paper focuses on information interoperability and cross domain information
sharing controlling communication systems for Future Internet network and services
approach. The extensible, reusable, common and manageable information inference
plane is critical for this deployment [7].
The novelty aspect of this approach relies on the fact that high level infrastructure
representations do not use resources when they are not being required to support or
deploy services. We optimize resources using this approach by classifying and
identifying, by semantic descriptions in a knowledge-based fashion way what resources
need to be used. Thus dynamically the service composition is executed and service
deployed by result of knowledge–based analysis.
Organization of this paper is as follows: Section II presents challenges for a
knowledge–based future Internet where information exchange occurs to support
composed service creation and delivery. Section III introduces our knowledge
engineering approach in a form of meta-ontologies facilitating information
interoperability and a demonstrator supporting the inference approach. Section IV
presents the summary and outlook and finally some relevant references used in this
paper are listed.
2. Challenges in a Knowledge-Based Future Internet
Taking a broad view of state of the art and current development of data link
interactions and converging communications networks as reference studied in this
paper, many of the problems present in current Internet interoperability are generated
by interoperability problems, we identify three persistent problems:
1. Users are offered relatively small numbers of Internet services, which they can not
personalise to meet their evolving needs; communities of users can not tailor
services to help create, improve and sustain their social interactions;
2. The Internet services that are offered are typically technology-driven and static,
designed to maximise usage of capabilities of underlying network technologies and
not to satisfy user requirements per se, and thus cannot be readily adapted to their
changing operational context;
3. Network operators cannot configure their networks to operate effectively in the
face of changing service usage patterns and rapid networking technology
deployment; networks can only be optimised, on an individual basis, to meet
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
specific low-level objectives, often resulting in sub-optimal operation in
comparison to the more important business and service user objectives.
As the move towards convergence of communications networks and a more extended
service-oriented architecture design gains momentum worldwide facilitated mainly by
pervasive deployment of Internet protocol suites, VoIP is a clear example of this, the
academic research community is increasingly focussing on how to evolve networking
technologies to enable the “Future Internet”. In this sense we believe that addressing
evolution of networking technologies in isolation is not enough; instead, it is necessary
to take a holistic view of the evolution of communications services, their societal
drivers and the requirements they will place on the heterogeneous communications
infrastructure over which they are delivered [8][9].
By addressing information interoperability challenge issues, Internet systems must be
able to exchange information and customize their services. So Future Internet can
reflect changing individual and societal preferences in network and services and can be
effectively managed to ensure delivery of critical services in a services-aware design
view with general infrastructure challenges.
3. Knowledge Engineering Approach – Semantic Annotation
An activity running currently is the composition of data models for enabling
information management control. It focuses in the semantic enrichment task of the
management information described in both enterprise and networking data models with
ontological data to provide an extensible, reusable, common and manageable inference
plane in the Future Internet systems.
The proposed knowledge-based approach provides tools to integrate user data with
the management service operations, and offers a more complete understanding of user’s
contents based on their social relationships and hence, a more inclusive governance of
the management of resources, devices, networks, systems and services for promoting
the integrated management with formal information models including social aspects.
This approach is to use ontologies as the mechanism to generate a formal description,
which represents the collection and formal representation for network management data
models and endow such models with the necessary semantic richness and formalisms
to represent different types of information needed to be integrated in network
management operations. Using a formal methodology the user’s contents represent
values used in various service management operations, thus the knowledge-based
approach over the inference plane [7] aims to be a solution that uses ontologies to
support interoperability and extensibility required in the systems handling end-user
contents for pervasive applications [10].
3.1 Service and Network Management meta-Ontology Modelling
The meta-ontology approach introduced in this section integrates concepts from the
IETF policy standards [11][12] as well as the TM Forum SID model [13][14]. In this
paper important classes that were originally defined in the IETF, SIM and DEN-ng
models will be cited and implemented as such, some other extended or adapted for
communication services adaptability, for more details see [15]. The meta-model defines
a set of interactions between the Context Data, Pervasive Management, and
Communications Systems Domains in order to define relationships and interactions
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
between the classes from the information models on these three different domains. The
meta-Ontology construction process, which is a four-phase methodology is result of
formal study to build up ontologies contained and studied from [16][17].
The formal language used to build the set of ontologies is the web ontology language
(OWL) [18][19], which has been extended in order to apply to pervasive computing
environments; these additional formal definitions act as complementary parts of the
lexicon. The formal descriptions about the terminology related with management
domain are included to build and enrich the proposal for integrating network and other
management data with context information to more completely define the appropriate
management operations using formal descriptions. The proposed meta-Ontology model
integrates concepts from policy-based management systems [20][21] to define a
context-aware system that is managed by policies, which is an innovative aspect of our
research work. Figure 2 shows the Ontology high level representation. The image
represents the integration of the entity information classes related to the management
operation class through the event class. The Event class interacts with other classes
from different domains in order to represent context information. Note that only the
InfoEntity class from the context information model domain and the Event class from
the service management domain are shown.
Context Data Domain
Ontology-Based User
Context Information Model Soperator
ManagementInfoApp
Policy Smanager
DataModel
Noperator
Condition Event Person
Action Info Entity Nmanager
Task Scheduled
PolicyApplication
Object Deduced
ManagedEntity
Ontology-Based Service
PolicyDecisionPoint Interactions
Place Application
PolicyDecisionMaking
Network
Ontology-Based
Ontology-Based Resource
Management Operations
Policy Information Model
Indoor
Pervasive Communications
Outdoor
Management Systems
Domain Domain
Figure 2. View of InfoEntity Integration in Service-Oriented Management Systems - by Domains.
This representation simplifies the identification of interactions between the
information models. These entity concepts, and their mutual relationships, are
represented as lines in the figure. The InfoEntity class forms part of an Event class, and
then the Event govern the policy functionality of a Managed Entity by taking into
account context information contained in Events. This functionality to enables context
to change the operation requested from a pervasive service or application, and is
represented as interaction between Event and InfoEntity.
The meta-Ontology model is driven by a set of pervasive service management use
cases that each requires service lifecycle policy-based management architecture as
represented in figure 3. The service composition and its model representation contains
the service lifecycle operations, as depicted in figure 3. In this figure, service
management operations, as well as the relationships involved in the management
service lifecycle process, are represented as classes. These classes will then be used, in
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
conjunction with ontologies, to build a language that allows a restricted form of
English to be used to describe its policies. To do so context information is underlayed
in such relationships, which a correspondence with activities called “events” could be
so related to context information.
Figure 3. Control and Representation of Service Lifecycle Interactions - UML model.
The meta-model is founded on information models principles for context
information and policy management promoting an integrated management, which is
required by both pervasive as well as autonomic management applications. The
combination of context-awareness, ontologies and policy-driven services motivates the
definition of a new, extensible, and scalable semantic framework for the integration of
these three diverse sources of knowledge to realize a scalable management platform.
From the Figure 3, the Service Editor Service Interface acts as the application that
creates the new service. Assume that the service for deploying and updating the service
code in certain network nodes has been created. This result in the creation of an event
named “aServiceOn”, which instantiates a relationship between the Application and
Maintenance classes. This in turn causes the appropriate policies and service code to be
distributed via the Distribution class as defined by the “aServiceAt” aggregation. The
service distribution phase finds the nearest and/or most appropriate servers or nodes to
store the service code and policies, and then deploys them when the task associated
with the “eventFor” aggregation is instantiated. When a service invocation arrives, as
signalled in the form of one or more application events, the invocation phase detects
these events as indication of a context variation, and then instantiates the service by
instantiating the association “aServiceStart”. The next phase to be performed is the
execution of the service. Any location-specific parameters are defined by the
“locatedIn” aggregation. The execution phase implies the deployment of service code,
as well as the possible evaluation of new policies to monitor and manage the newly
instantiated service.
Monitoring is done using the service consumer manager interface, as it is the result
of associations with execution. If maintenance operations are required, then these
operations are performed using the appropriate applications, as defined by
the“aServiceOn” aggregation, and completed when the set of events corresponding to
the association “whenServiceOff” is received. Any changes required to the service code
and/or polices for controlling the service lifecycle are defined by the events that are
associated with the“whenServiceNew” and “aServiceChange” associations.
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
The service management operations are related to each other, and provide the
necessary infrastructure to guarantee the monitoring and management of the services
over time. The UML design shown in figure 3 captures these relationships, thus the
pervasive service provisioning and deployment is on certain manner assured to provide
service code and policies supporting such services to the service consumers. The
coding process for formalizing the ontology follows the definitions and specifically the
description contained in the filed “relation” from the definitions provided in this
chapter. It is very important to understand the real sense of the “relation” field, as this
represents how the descriptions and concepts are used to build and enhance the
proposed ontology.
The base or this ontology code is included in [10]. However this section contains
partial description about some of the concepts more relevant when ontology is created,
descriptions, definitions and integration of concepts are also included and described.
Figure 4 shows the OWL grammar section of the ontology that describes the
InfoEntity and its corresponding domain elements. It is a description of an object class
for representing an InfoEntity, and represents the simplest definition and relationships
in sense of disjointness to the Application, Place, Person, DataBaseIM and Task classes.
The disjoints represent a semantic tool for filtering the seek of information, thus the
objects classes including disjoints can be easily identified to be considered or not as
part of the knowledge that is being seek.
Condition Class
Figure 4. Event InfoEntity Description using OWL Grammar Representation.
The use of XML, and the resulting use of RDF extensions, aimed to improve the
expressiveness of ontology languages. The RDF-Schema (RDFS) [22] language
emerged as a set of extensions to provide increased semantics of RDF by providing
basic ontological modelling primitives, like classes, properties, ranges and domains.
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
One of the advantages of using OWL to express the ontology is to provide a
number of tools for parser and text editors. This enables new adopters to use a tool or
set of tools that is best suited to their needs. OWL is used to define the set of concepts
and constraints imposed by the information model over which it defines instances.
3.2 Functional Architecture Approach
A functional diagram for functional architecture supporting the meta-Ontology
with its functional components as initial approach is depicted in Figure 5. This diagram
shows the interactions between the main components. The architecture controls with a
certain high level domain-based view, how service composition is performed, thus the
control of the behavior is considering like added value functionalities using high level
and formal representations expanding operations in other service application domains.
The architecture presented is using information to manage service operation and
instructions with ontology-based information models using semantic mechanism.
Based on previous implementation experience [23][24] this architecture allows
adaptability and dynamism to current and future Internet services with the advantages
of incorporating context from users and applications in form of events.
This functional approach offer more functionality than standard service-aware
management approaches, as they cannot orchestrate system behaviour using knowledge
from business, network, and other constituencies. By representing these data in a
formal manner, data can be integrated, and the power of machine-based learning and
reasoning can be more fully exploited. In this approach we translate data from a device-
specific form to a device- and technology-neutral form to facilitate its integration with
other types of information. The key difference in this architecture, compared with a non
ontology-based enabled, is the use of semantic information to guide the decision-
making process.
Figure 5. Semantic-Based Service Control Engine - Functional Architecture.
Observation is translated to events to be co-related with the system’s behaviour and
its activity and then learned to make the system react when the same event occurs in the
future. Co-related events trigger and control each set of independent related events thus
certain level of autonomy is achieved. The service composition process involves
analyzing the triggering events expressed in an appropriate interoperable language via
service coordination and decision-making integration, and matches them to service
management and control level available [25] with the difference of this component
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
using semantic descriptions to co-relate events with particular kind of conflicts that
must be identified and evaluated. A detailed Semantic-Based Service Control Engine
(2SCE) and its components as part of Functional Architecture is out of the scope of this
paper, however implementation results are being analyzed and interaction between
different domain events tested successfully.
4. Conclusions
In the future Internet high demands of information interoperability to satisfy service
composition requirements being controlled by diverse, heterogeneous systems make
more complex perform system management operations. Thus the approach introduced
in this paper emerges as an alternative to solve part of those information
interoperability complex requirements in the Future Internet of networks and services.
We have studied and demonstrated how formal representation of service and
networks information facilitates information interoperability in service composition
and management processes. Remaining research challenges regarding information
model extensibility and information dissemination exist and would be conducted to
conclude implementations, experiments composing services.
This paper makes references to formal and theoretical foundations for the
development of interoperability of information by using knowledge engineering
techniques, semantic aggregation. Information interoperability is a crucial requirement
in Future Internet. Their implications for networks and services is still and open issue
for research and interoperability and information exchange must be validated via direct
industrial investment, and roll out real integrated test beds to trial new network
infrastructures (potentially overlay networks or new inference plane infrastructures) for
services.
Acknowledgements
This research activity is partially funded by Science Foundation Ireland (SFI), grant
08/SRC/I1403 FAME-SRC (Federated, Autonomic Management of End-to-End
Communications Services - Scientific Research Cluster). Project partners participate
actively in EU Projects and FIA Initiative.
References
[1] D. Clark, et al., “NewArch: Future Generation Internet Architecture”, NewArch Final
Technical Report, http://www.isi.edu/newarch/
[2] M. Blumenthal, D. Clark, “Rethinking the design of the Internet: the end to end arguments vs.
the brave new world”, ACM Transactions on Internet Technology, Vol. 1, No. 1, Aug. 2001
[3] A. Feldmann, “Internet clean-slate design: what and why?,” ACM SIGCOM Computer
Communication Review, Vol. 37, Issue 3, 2007
[4] NSF-funded initiative to rebuild the Internet, http://www.geni.net/
[5] Clean Slate program, Stanford University, http://cleanslate.stanford.edu
FIA Session “Linked Data in the Future Internet” - Position Paper
FIA Assembly, Ghent, 16 December 2010
[6] Architecture Design Project for New Generation Network, http://akari-
project.nict.go.jp/eng/index2.htm
[7] J. Strassner, M. Ó Foghlú, W. Donnelly, N. Agoulmine, “Beyond the Knowledge Plane: An
Inference Plane to Support the Next Generation Internet”, IEEE GIIS 2007, 2-6 July, 2007.
[8] Irish Future Internet Forum, http://www.futureinternet.ie
[9] SFI-SRC FAME - Scientific Research Cluster: "Federated Autonomic Management of End
to End Communications Sesrvices, http://www.fame.ie/
[10] M. Serrano, J. Strassner and M. ÓFoghlú, “A Formal Approach for the Inference Plane
Supporting Integrated Management Tasks in the Future Internet“ 1st IFIP/IEEE ManFI
International Workshop, In conjunction with 11th IFIP/IEEE IM2009, 1-5 June 2009, at
Long Island, NY, USA.
[11] Westerinen, A.; Schnizlein, J.; Strassner, J. “Terminology for Policy-Based Management”.
IETF Request for Comments (RFC 3198). November 2001.
[12] Moore, E.; “Policy Core Information Model-Extensions”. IETF Request for comments
(RFC 3460), January 2003. http://www.ietf.org/rfc/rfc3460.txt
[13] TMF, “The Shared Information and Data Model – Common Business Entity Definitions:
Policy”, GB922 Addendum 1-POL, July 2003.
[14] SID - Shared Information Data model.
http://www.tmforum.org/InformationManagement/1684/home.html
[15] Serrano, J.M. “Management and Context Integration Based on Ontologies for Pervasive
Service Operations in Autonomic Communication Systems”, PhD Thesis, UPC 2008.
[16] Horridge, M., Knublauch, H., Rector, A., Stevens, R., Wroe, C., “A Practical Guide to
Building OWL Ontologies using the Protégé-OWL Plugin and CO-ODE Tools Edition 1.0”
Manchester University, August. 2004.
[17] Gruber T., Towards “Principles for the Design of Ontologies Used for Knowledge Sharing”,
International Journal of Human-Computer studies, 43 (5/6): 907 – 928. 1995.
[18] OWL Ontology Web Language. http://www.w3.org/2004/OWL
[19] OWL-s http://www.daml.org/services/owl-s/
[20] M. Sloman. “Policy Driven Management for Distributed Systems”. Journal of Network and
Systems Management, 1994
[21] Strassner, J., “Policy Based Network Management”, Morgan Kaufmann, ISBN 1-55860-
859-1, 2004.
[22] W3C Website. http://www.w3c.org/rdf
[23] Serrano, J. Martín; Serrat, Joan; Strassner, John, “Ontology-Based Reasoning for
Supporting Context-Aware Services on Autonomic Networks” 2007 IEEE/ICC
International Conference on Communications, 24-28 June 2007, Glasgow, Scotland, UK.
[24] J. Martín Serrano, Joan Serrat, Declan O’Sullivan; “Onto-Context Manager Elements
Supporting Autonomic Systems: Basis & Approach”. 1st IEEE MACE International
Workshop as part of ManWeek 2006. 23 - 27 October 2006, Dublin, Ireland.
[25] Schönwälder, J. Straub F. “Next Generation Structure of Management Information for the
Internet”. In Proceedings of the 10th IFIP/IEEE DSOM International Workshop, Zürich,
1999.