Services engineering

Infrastructure services are an important technological sector with a variety of aspects. For at least a decade the models of managed application- or storage-hosting services, contract-based services, and information technology outsourcing have been continuously growing in the enterprise computing sector. More recently the emergence of cloud computing as a popular alternative to owning information technology infrastructure has brought the computing-as-a-service model into the mainstream. Nowadays, anyone can ask and rapidly be allocated a virtual machine instance or storage volume to use on a pay-as-you-go basis.

Infrastructure services can take two forms – infrastructure provided as a service (IS) or services provided on an infrastructure (SI). Examples of the former approach are managed application-hosting services, information technology outsourcing on the provider’s infrastructure, and cloud computing. The common theme across these cases is that the service being provided is the use of managed infrastructure. Examples of the latter approach include consulting engagements providing value to the client infrastructure by better aligning it to its business goals.

The field of infrastructure services raises a variety of research challenges. First, in both the IS and SI models there is a need for effective large-scale management of heterogeneous provider-owned or customer-owned infrastructure. Concerns such as rapid troubleshooting, change management, allocation and management of resources on rapidly-varying workloads and so on require up-to-date understanding of system state and configuration and mapping of the infrastructure to the business goals it supports. Second, infrastructure service providers have a need for well-defined processes and tools that help them provide effective services in a variety of client environments. Especially in the SI model there is thus a need for flexible service engineering to adapt to widely varying client needs.

The project aims at advancing the state of the art in all these areas by performing research in model-driven tools for rapidly prototyping service-driven applications and for configuring and managing heterogeneous environments; large-scale configuration management repositories for service-oriented infrastructures; cloud computing service delivery models.

Related publications:

• Adapting Data-Intensive Workloads to Generic Allocation Policies in Cloud Infrastructures, Giannis Kitsos, Antonis Papaioannou, Nikos Tsikoudis, and Kostas Magoutis, In Proceedings of IEEE/IFIP Network Operations and Management Symposium (NOMS 2012), Maui, Hawaii, April 16-20, 2012.
• Built-to-Order Service Engineering for Enterprise IT Discovery, Nikolai Joukov, Murthy Devarakonda, Kostas Magoutis, Norbert Vogl, in Proceedings of IEEE International Conference on Services Computing (SCC 2008), Hawaii, vol. 2, pages 91-98, July 2008.
• Galapagos: Model-Driven Discovery of End-to-End Application-Storage Relationships in Distributed Systems, Kostas Magoutis, Murthy Devarakonda, Nikolai Joukov, Norbert Vogl, in IBM Systems Journal, vol. 52, no. 4/5, pages 367-378, 2008.

ScaleWorks is a Marie Curie IEF Project, Contract No. PIEF-GA-2009-237677

Project funded by the European Commission 7th Framework Program.