Data Spaces are a rapidly evolving concept that enable sovereign and secure data sharing. A Data Space consists of multiple building blocks and follows a well-de!ned governance and security framework. One of these key building blocks is a catalog for data discovery, which allows participants to publish and !nd data, services, and other artifacts in a compliant and trustworthy manner. Additionally, many Data Space standards, such as Gaia-X and Veri!able Credentials, utilize Semantic Web technologies as a data modeling language. With piveau-X, we introduce a catalog solution for Data Spaces that elevates the Semantic Web, enabling the seamless application of formal Data Space speci!cations. Our solution employs a Triplestore as the primary database and SHACL as the data modeling language, ensuring maximum "exibility in managing data and metadata structures while also facilitating the dynamic reuse and integration of existing ontologies. Furthermore, piveau-X natively integrates with identity and attestation services, making it a comprehensive solution for creating Data Space catalogs. Its built-in SPARQL and Elasticsearch-powered search and query capabilities allow for precise browsing of data artifacts via APIs or the web frontend. We successfully validated piveau-X's practical applicability in the Gaia-X research project POSSIBLE through !ve real-world use cases.
1. Introduction
Data Spaces represent a thriving and widely adopted concept that facilitates sovereign and secure data
and service sharing. Unlike centralized data integration mechanisms, Data Spaces keep data at its source
without transforming it into a uni!ed format. Integration occurs semantically via a shared vocabulary
and metadata model. Furthermore, the concept introduces a business and legal layer that ensures trust,
interoperability, portability, and data sovereignty. [
A variety of technical and organizational building blocks are established to instantiate Data Spaces.
One of the central building blocks is a catalog that enables the publication and discovery of data, services,
o#erings, and other artifacts. Another key building block is a trust framework, which ensures the
integrity and authenticity of o#ered resources by leveraging trust anchors, compliance rules, and digital
identities. [
We propose that implementing the core concept of Data Spaces and their central building blocks
can be achieved through the full adoption of Semantic Web technologies, such as RDF, SHACL, and
SPARQL. While these technologies are already applied in various "avors and applications of Data
Spaces, they are not yet utilized to their full potential. In this demo paper, we introduce our Semantic
Web-based catalog solution, piveau-X, which aims to maximize the use of Semantic Web technologies
while ensuring compliance and trust. Our solution o#ers the following core features: (1) the use of a
Triplestore as the primary database, enabling high "exibility and rich query capabilities via SPARQL,
(2) a "exible and dynamic metadata model con!guration based on SHACL, combined with full-text
search powered by Elasticsearch, (3) a user-friendly web frontend for !ltering and searching assets, and
(4) direct integration with identity and attestation services. Our solution is built on the open-source
catalog software piveau [
This section provides an overview of the foundational concepts that underpin our research, including, Gaia-X, Decentralized Identi!ers (DIDs) and Veri!able Credentials (VCs), and Data Spaces Support Centre (DSSC).
Gaia-X is a European initiative for an interoperable, federated data infrastructure that ensures
data sovereignty for its participants. By establishing a framework built on European principles of
data security, GDPR compliance, transparency, and interoperability, Gaia-X enables organizations to
share data securely and under rules that preserve control for data owners [
As part of this framework, DIDs are used to establish digital identities for organizations and VCs
are used for authentication and authorization purposes. DIDs are W3C standardized self-sovereign
identi!ers that allow entities to autonomously create, manage, and update cryptographically veri!able
digital identities without centralized authorities [
The DSSC is an initiative funded by the European Commission that aims to identify and de!ne
standards, best practices, technologies, and tools to build and foster European Data Spaces. The DSSC
supports public sector entities and businesses in creating and managing these Data Spaces, thereby
accelerating the digital transformation across various domains. [
Piveau, a comprehensive and scalable data management catalog, was extended to meet the
requirements of a Data Space catalog. The solution is based on Semantic Web technologies, utilizes a Triplestore
as its primary database, and Elasticsearch as its indexing and search service. Furthermore, piveau is
primarily designed to support Open Data platforms. [
In addition to these foundational technologies, several related works have explored practical
implementations of Gaia-X compliant catalogs. The Eclipse Cross Federation Services Components (XFSC)
Catalog was developed as a reference implementation of the Gaia-X federated cloud service architecture.
It serves as a directory of services and providers, managing metadata records, schemas, and VCs while
enabling service searchability, comparability, and data sharing. To support domain-speci!c
requirements, it allows the submission of custom metadata schemas de!ned using OWL, SHACL, or SKOS.
Furthermore, it performs its own syntactic, cryptographic, and schema veri!cation checks on VCs,
as opposed to using the Gaia-X Digital Clearing House (GXDCH). Notably, the XFSC Catalog avoids
storing RDF data altogether, instead relying on openCypher’s labeled property graph (LPG) data model
for storing and searching data. [
Aster-X, created under the French Gaia-X Federation Services initiative, also makes use of the Eclipse
XFSC. Developed as a Gaia-X Data Space demonstrator, the Aster-X catalog showcases functionalities
such as Self Sovereign Identity (SSI) wallets, SSI-based permissions, automated contract negotiation,
data exchange, and direct integration with the GXDCH for compliance veri!cation. Similar to the XFSC
Catalog, Aster-X stores data using openCypher’s LPG data model. [
Unlike XFSC-based Gaia-X catalogs, Prometheus-X implements its own open-source building blocks
to facilitate data sharing with the Gaia-X ecosystem. More speci!cally, Prometheus-X comprises a set
of core services to facilitate participant onboarding, service discovery, contract negotiation, consent
management, and direct integration with the GXDCH for compliance veri!cation. [
Pontus-X is the Gaia X web3 ecosystem that uses Distributed Ledger Technology (DLT) to store
records of data sharing contracts, settlements, and usage rights, and enforces them via smart contracts.
Pontus-X makes use of the Ocean Protocol to store metadata on-chain, connects the metadata to a
database, and integrates the GXDCH for compliance veri!cation. Furthermore, the Pontus-X catalog
focuses on publishing a wide variety of software services, and thus does not support extensible data
schemas. [
The primary objective of piveau-X is to provide an out-of-the-box solution for the Data Space building
block Publication & Discovery [
In the following sections, we present the detailed design decisions and overall architecture of piveauX. Figure 1 illustrates the architecture, depicting all components and services. From a high-level perspective, we divide our architecture into two realms: the Data Space Realm and the piveau-X Realm.
The Data Space Realm provides essential Data Space services required for a functional catalog. A Decentralized Identity Service manages the digital identities of participants and authorizes their interactions within the Data Space. An Attestation Service veri!es cataloged resources against compliance rules and digitally signs them. Data Models de!ne the format, structure, and constraints of resource descriptions. Finally, an independent Veri!cation Service ensures the integrity of published resources.
The piveau-X Realm, which consists of the catalog, is divided into three layers: Con!guration, Databases and Integration, and Main Services.
The Con!guration layer establishes a connection to the Data Space Realm. SHACL !les serve as a direct representation of the Data Models, de!ning the supported data structures of the catalog. The Identity Keys are cryptographic private keys linked to the Identity Service and, consequently, to a speci!c participant. The Pro!le represents the general runtime con!guration of the catalog.
The Databases and Integrations layer comprises two main databases: a Triplestore for managing and storing RDF and an Elasticsearch search server for enabling high-performance full-text search. This layer also includes a Signing Service, which is responsible for cryptographically signing resources using the provided private keys. This functionality is o"ered as an independent service to enable self-sovereign key management and signing, allowing on-premises deployment when using a central catalog instance.
The Main Services layer consists of the core services. The Search Service provides an interface to the Elasticsearch server, the Repository functions as middleware for the Virtuoso Triplestore, and the Web Frontend delivers a user-friendly interface. Table 1 illustrates how the various design aspects of piveau-X align with the technical pillars of the Main Services Databases and Integrations Configuration
Provider and Consumer Catalog
Web Frontend Search Service
Repository Elasticsearch Signing Service SHACL Files Identity Keys piveau-X Realm
Virtuoso Base Config
Decentralized Identity Service
Attestation Service
Data Models Verification Service Data Space Realm Data Space Blueprint. As the foundation for our architecture, we chose to build upon an existing, mature catalog solution: piveau. While piveau is already a comprehensive and scalable data management catalog based on Semantic Web technologies, it can be extended to meet the requirements of a Data Space catalog.
The piveau-X catalog instantiates this architecture in the Gaia-X Data Space domain1. In this case,
the Data Space Realm consists of a Decentralized Identity Service that provisions and updates an
organization’s DIDs, DID Documents, and associated certi!cates used to identify organizations and for
secure signing and veri!cation of VCs [
In the Con!guration layer, the piveau-X catalog leverages SHACL shapes to de!ne custom metadata models across various domains. In this context, the Gaia-X Legal Participant, Service O"ering, and Data Resource models are each represented as dedicated SHACL !les.
Beyond de!ning the metadata schema, these SHACL !les also specify how RDF metadata is indexed
and stored as JSON in Elasticsearch. For example, the Gaia-X Core Ontology de!nes the attribute
gx:providedBy for a Service O"ering resource, indicating the Legal Participant responsible for
providing the service [
how gx:providedBy is indexed to JSON, using the piveau StandardKeyword indexing function and setting the indexed attribute name to provided_by. This dual representation of RDF and indexed JSON data enables the piveau-X catalog to support robust semantic querying via SPARQL and deliver e#cient full-text search capabilities.
Once the piveau-X catalog is con!gured, users register their organizations in the Gaia-X ecosystem
as Legal Participants, with the resulting resources stored in the Databases and Integrations layer. To
do so, users send an API request to the catalog whose payload conforms to the Legal Participant
SHACL shape, which includes !elds such as the organization’s name, description, address, and Legal
Registration Number (i.e., VAT ID). This information is then forwarded to the Gaia-X Notary Service,
which returns a Legal Registration Number VC upon verifying the organization’s legal existence. Next,
a Legal Participant VC is composed and signed by a separate signing service developed speci!cally for
the Gaia-X Data Space, using normalization and signing algorithms de!ned by Gaia-X and con!gured
with private keys associated with the Legal Participant’s corresponding DID [
As illustrated in Figure 2, Providers (Legal Participants o"ering services) publish Service O"erings by sending an API request conforming to the Service O"ering SHACL shape. This payload includes !elds like the Service O"ering’s name, description, contract policy, the providing organization, and associated Data Resources. A Service O"ering VC and associated Data Resource VCs are created, signed, and submitted along with the organization’s existing Legal Participant and Legal Registration Number VCs to the GXDCH. If compliant, the GXDCH issues a Compliance VC, after which the Service O"ering resource and all related VCs are stored in the piveau-X catalog, maintaining Gaia-X compliance integrity.
In the Main Services layer, consumers can view catalog resources via a web-based interface that provides an overview of all Legal Participant and Service O"ering resources, with Service O"erings further categorized based on whether they include aggregated Data Resources. From the Legal Participant details page, users can view the resource attributes and inspect the corresponding Legal Participant VCs, Legal Registration Number VC, and the Compliance VC returned by the GXDCH. Similarly, the Service O"ering details page displays the Service O"ering’s attributes along with any associated Data Resources. Users can also view the corresponding Service O"ering and Data Resource VCs and con!rm their Gaia-X compliance via the Compliance VC. This interface thereby ensures transparency into the complete lifecycle of a resource—from creation to veri!cation and storage.
We successfully validated piveau-X within the Gaia-X project POSSIBLE, whose objective is to demonstrate sovereign data value creation with real-world partners2. It integrates Gaia-X artifacts and other tools into a reusable toolbox to facilitate the bootstrapping and maintenance of a Data Space. Our solution is successfully integrated within the Data Space base layer of POSSIBLE, and is complemented by an end-user portal and the Eclipse Dataspace Components (EDC) Connector3 for actual data exchange.
The following !ve pilot use cases leverage the POSSIBLE platform to provide Gaia-X-compliant Service O"erings and Data Resources: an interoperable messenger for public administration, sharing of non-public geodata, a skill analytics service, an AI-driven data product designer, a recommender for learning material, and a secure data-sharing service. For all stakeholders—both providers and consumers—of these use cases, we created and signed compliant Legal Participants in piveau-X, with the POSSIBLE portal acting as the gateway. Service O"erings are provided in a similar manner via the POSSIBLE portal and are cross-linked to the issuing participant within our catalog. POSSIBLE integrates with EDC connectors for data exchange and Open Digital Rights Language (ODRL) for describing policies. To manage the required information, such as EDC identi!ers, we provided a SHACL-de!ned
3https://github.com/eclipse-edc/Connector (a) Landing Page (b) List of Data Resources (c) Detail Page of a Data Resource (d) Detail Page of a Legal Participant data model that seamlessly integrates into the GXDCH. The entire solution is deployed in a Kubernetes cloud, and the use case evaluation was successfully completed. Figure 3 shows some impressions of the user interface. Additionally, we developed a Veri!cation Service4 that wraps a real Gaia-X veri!cation library to verify the technical compliance and correctness of signed artifacts.
Thus, the project has demonstrated that piveau-X can be e"ectively applied in real-world Data Space scenarios. Furthermore, the integration of Semantic Web technologies was instrumental in integrating the Gaia-X data model and accommodating use-case speci!c requirements. After all, the Gaia-X data model itself is based on SHACL and RDF.
In this paper, we presented our Data Space catalog implementation, piveau-X. Our solution enables the deployment and con!guration of a catalog for a Data Space, allowing for the dynamic customization of the data model based on SHACL. As a result, the catalog can be tailored to the speci!c needs and requirements of a given Data Space. Furthermore, piveau-X provides direct integration points with identity and attestation services, making it a comprehensive solution for managing Data Space o"erings. Furthermore, we successfully evaluated the catalog in real-world use cases within the Gaia-X research project.
In summary, our solution addresses challenges across all three technical building blocks of the DSSC blueprint. The direct use of Semantic Web technologies for de!ning data models enhances data interoperability. Integration with identity and attestation services ensures data sovereignty and trust. Finally, the core functionality that enables the discovery of o"ering descriptions allows data value creators to identify and reuse suitable data. Additionally, our approach aligns with the "Data Discovery"
challenge from the W3C Dataspaces Community Group, as the Semantic Web already provides a well-de!ned speci!cation for describing, linking, and discovering data.
We are working to further extend and enhance piveau-X to reach production-grade maturity. piveauX is a core component of a data trading platform developed in the scope of the European funded project PISTIS5. PISTIS emphasizes the monetization of proprietary data assets in a trustworthy and controlled environment, and is also following the core aspects of the DSSC blueprint. The project will demonstrate its outcomes in three large-scale demonstrators in mobility, energy and automotive domains. We are re!ning the generic data model design of piveau-X to enable even greater $exibility for use in PISTIS and beyond Gaia-X. PISTIS uses DLT for identity management and as a trust anchor. We are aiming to integrate this as another type of Attestation Service to demonstrate the pluggable trust mechanism of piveau-X.
Additionally, we are working to introduce extended lifecycle management for data, supporting multiple roles and states of o"erings throughout the publication process. Finally, we intend to integrate a direct feature for data exchange based on the Data Space Protocol6.
This work has been funded by the Federal Ministry for Economic A"airs and Climate Action (BMWK) under grant no. 68GX21009B (“POSSIBLE: Phoenix open software stack for interoperable engagement in dataspaces”) and by the European Union under the Grant Agreement 101093016 ("PISTIS: Promoting and Incentivising Federated, Trusted, and Fair Sharing and Trading of Interoperable Data ASsets").
During the preparation of this work, the authors used ChatGPT solely for grammar and spelling checks. After applying the tool’s suggestions, the authors reviewed and edited the text and accept full responsibility for the published content.
6https://docs.internationaldataspaces.org/ids-knowledgebase/dataspace-protocol
• Documentation: https://doc.piveau.io/hub/piveau-x • Live Demo: https://possible.fokus.fraunhofer.de