Semantic-based management of federated infrastructures for future internet experimentation

Sharing and granting access to geographically dispersed resources is the underlying concept in the field of Distributed Computing. Associated with this, considerable efforts have been spent on architectures to manage heterogeneous resources across multiple administrative domains. Such an environment is referred to as a resource federation. A specific field of application is the execution of experiments in the context of Future Internet research. Given the scale, complexity and heterogeneity of the Internet, experimental validation is carried out within large-scale, distributed test environments. As a result, several independent testbeds have been established, with accompanying protocols to support the experiment life cycle across sites. An important challenge that arises in this context is the exchange of information about the provided resources with their types and characteristics. Existing work rests upon certain interfaces and syntactic data models with arbitrary extensions and identifiers, which aggravate the management of heterogeneous resources across autonomous testbeds. This thesis introduces an approach that is founded on well-defined semantic information models and architectural abstraction to address these issues. Based on mechanisms that have their origins in Semantic Web research, the use of semantically annotated graphs allows for automatic reasoning, linking, querying and validation of heterogeneous data. The main contributions of the research conducted for this thesis are the definition of an ontology for the life cycle management of resources in federated infrastructures, a corresponding semantic- and microservice-based architecture for interface abstraction, and a proof-of-concept implementation. The work has been validated within several European Future Internet research projects and testbeds. Further, a performance evaluation has been carried out and contributions to relevant standardization activities have been made. In general, the approach forms a basis for further work in the context of distributed resource management in federated en- vironments, such as Intercloud and Edge Computing approaches, multidomain Software Defined Networks (SDNs), and the Internet of Things (IoT).

Author dc:creator

• Willner, Alexander

Advisor dc:contributor.advisor

• Magedanz, Thomas

Licence dc:rights.uri

• https://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en

Identifier URI

http://dx.doi.org/10.14279/depositonce-5012

OAI identifier oai:identifier

oai:depositonce.tu-berlin.de:11303/5327