EOSC Accounting Service

EOSC Accounting for Services is a platform designed to collect, aggregate, and exchange metrics efficiently across infrastructures, providers, and projects. It provides a REST API for ingesting data from diverse resources, stores the information in a database, and aggregates it for analysis. It allows clients to interact with the system and retrieve accounting data for specific time periods. Access to all API resources is secured through client authentication, ensuring that sensitive information is protected. Core Capabilities Efficient collection, aggregation, and exchange of metrics REST API for input from heterogeneous resources Database-driven storage Reports containing aggregated metric information User-friendly interface for querying accounting information Secure access through authenticated clients The service offers a structured framework for managing accounting data by projects, providers, or services/installations, with different administrative roles such as Project Admin, Provider Admin, and Service/Installation Admin, each holding distinct responsibilities. This document presents a proof of concept (PoC) for leveraging the EOSC Accounting service to support Responsible Research Assessment (RRA) within the GraspOS project. It proposes a unified framework that integrates two complementary types of metrics: publication usage data (e.g., views, downloads, citations from repositories like TR-Dizin and OpenAIRE) and HPC resource usage data (e.g., CPU/GPU core hours, energy consumption from infrastructures like TRUBA). By combining these metrics, the EOSC Accounting service enables a multidimensional research evaluation model that captures not only the scholarly impact of research outputs but also the computational effort, efficiency, and environmental costs associated with their production. This approach supports more equitable, transparent, and Open Science-aligned assessment practices, addressing limitations of traditional citation-based metrics. The PoC outlines the technical implementation via the EOSC Accounting API, including metric definitions, hierarchical data structures, and reporting endpoints for both group and individual levels. It also identifies current operational challenges—such as the lack of project-based accounting in HPC systems—and proposes future steps for broader adoption, including improved metadata practices, persistent identifiers, and structural adaptations. This initiative demonstrates how infrastructure-aware metrics can transform research assessment, promoting fairness, sustainability, and better alignment with modern data-intensive research practices.