NORDUnet

This deliverable is the first of a series of three studies of cross-border service delivery and collaboration in EOSC planned by EOSC-Nordic. We look at EOSC from the perspective of the delivery chain and identify the different actors that belong to the EOSC delivery chain and map their roles and responsibilities, based on a number of service case studies. Future work will study existing cross-border collaboration models in the Nordic and Baltic countries (for publication in February 2021), and propose recommendations for future cross-border collaboration in the context of EOSC (for publication in August 2021).<br> A successful delivery chain involves many actors. Each actor plays a critical and unique role that is crucial to the delivery of the services. Becoming a part of the EOSC service ecosystem means dealing with an additional layer(s) of complexities and interconnections in the delivery chain, which is an aspect that should be planned and executed meticulously.<br> The work reported here is part of the Policies, legal issues, and sustainability effort of the EOSC-Nordic project to foster coordination of and connection between national initiatives at the policy level, in the context of EOSC. The deliverable will help policymakers stay informed about cross-border service delivery and service provision in the context of EOSC. The deliverable will also be input to the EOSC governance and architecture evolution.<br> In this report, we consider service case studies and illustrate that the service delivery chain for ICT services for research requires complex and careful technical and organisational orchestration, dependent on the interplay of many stakeholders, on existing national and international agreements outlining rights and responsibilities, and on tradition and human networks. We show that these relationships depend on the nature of the services, the community of users, and the basis for the funding of the services. We have seen also that these relations are especially complex in a cross-border service delivery environment.<br> We analyse different aspects of the service delivery chain, considering user communities, funding, operations and support, resource types, and governance. We show that all these aspects influence how the service may be offered and its relation to EOSC. We show that this takes many forms and is different for all of the six service cases considered. The cases have been chosen on the basis that they all reveal important aspects of the delivery chain, and for this reason, some of the cases have an above-average delivery chain complexity. Our analysis of the six services reveals that some of these services are generic and can be offered universally through EOSC, provided that funding to replenish consumable resources can be ensured. Other services that mainly provide access to data can be provided through EOSC, without additional funding, when the stakeholders have incentives to do so. Some services are national in scope and governance and may therefore be regarded as strategically important nationally. For such services, there might be policy considerations to take into account - in addition to the technical and workflow challenges. Finally, some services are specifically tailored towards a research community and as such integral to the research practices of that community. Incentives for the community to offer these services through EOSC may not be clear and should be assessed case by case.

Because of the static nature of the deployed optical networks, large energy wastage is experienced today in production networks such as Telecom networks. With power-adaptive optical interfaces and suitable grooming procedures, we propose the design of more energy efficient transport networks. Optical network re-configurations are performed by GMPLS node controllers according to monitored traffic information. The investigated energy reduction strategies are simulated on two large scale transport networks (DT17 and COST37). The results show that the energy savings obtained by these strategies depend on the variability of the carried traffic and the characteristics of the network topology. For medium-size DT17 network significant (more than 37%) power savings are achieved only with symbol-rate adaptation while less savings are achieved for modulation format adaptation. In case of pan-European COST37 network, for both symbol-rate and modulation format adaptations significant savings are obtained. Mixed adaptation (jointly performing symbol-rate and modulation format adaptations) used together with optical grooming allows up to 44% and 47% power savings in DT17 and COST37 networks respectively, close to the optimum power savings case 48,2% computed when the power follows exactly the traffic demand pattern.

The Tier-1 facility operated by the Nordic DataGrid Facility (NDGF) differs significantly from other Tier-1s in several aspects: It is not located at one or a few locations but instead distributed throughout the Nordic, it is not under the governance of a single organisation but but is instead build from resources under the control of a number of different national organisations. Being physically distributed makes the design and implementation of the networking infrastructure a challenge. NDGF has its own internal OPN connecting the sites participating in the distributed Tier-1. To assess the suitability of the network design and the capacity of the links, we present a model of the internal bandwidth needs for the NDGF Tier-1 and its associated Tier-2 sites. The model takes the different type of workloads into account and can handle different kinds of data management strategies. It has already been used to dimension the internal network structure of NDGF. We also compare the model with real life data measurements.

Interoperability of grid infrastructures is becoming increasingly important in the emergence of large scale grid infrastructures based on national and regional initiatives. To achieve interoperability of grid infrastructures adaptions and bridging of many different systems and services needs to be tackled. A grid infrastructure offers services for authentication, authorization, accounting, monitoring, operation besides from the services for handling and data and computations. This paper presents an outline of the work done to integrate the Nordic Tier-1 and 2s, which for the compute part is based on the ARC middleware, into the WLCG grid infrastructure co-operated by the EGEE project. Especially, a throughout description of integration of the compute services is presented.