RAISE distributed data visiting network as a service

RAISE is featured as project of the month on cordis (https://cordis.europa.eu/projects)) as a technology for a more transparent approach to sharing, visiting and processing data. The novelty of RAISE lays on the combination of technologies that transform the data sharing and processing procedure to a “data visiting" model. The RAISE system provides the infrastructure as a service for a distributed data storage and processing system that shifts the focus from “open data” to “data open for processing”, while adhering to the FAIR (Findable, Accessible, Interoperable, Reusable) principles. Rather than transferring data to local environments, the RAISE Central Hub dispatches the data users’ algorithms to the location where data is stored (RAISE Nodes). For each data processing and the corresponding result, RAISE issues Persistent Identifiers (PID) and registers the corresponding metadata to the RAISE blockchain.
Data owners can deploy their own nodes in their premises, keeping their datasets securely within their own infrastructure, never losing control or ownership. The nodes are securely connected and their status and remote execution task list is available on the RAISE Central Hub in real-time. The registries in the Hub contain metadata for every dataset, script, node, experiment and experiment result available in the Central Hub or in the distribute network of RAISE Nodes. The real data are never available for download by other users. Instead, shareable data samples with the same structure and format as the real dataset can be generated manually or automatically.

RAISE effectiveness has been measured and statistically significant results show that RAISE addresse the fears and concerns of researchers for:

• lack of recognition and acknowledgment for data reuse by others

• Misinterpretation and falsification

• Sensitive data protection

• Career benefits and visibility

• Be and feel in control of data usage when made open

How it works:

• Data owners can deploy their nodes in their premises, storing their data there.

• Datasets remain on-premises; only approved algorithms can run on them.

• Metadata of datasets are searchable to help researchers discover datasets.

• Only a sample (or synthetic version) of the dataset is downloadable for script development.

• Data owners can either create these downloadable samples manually or allow RAISE to automatically generate a synthetic dataset from the real dataset.

• Data owners can require access approvals, so that they have to approve requests by other researcher for processing their datasets.

• Data owners retain full control of who processes the data and can revoke access at any time.

• Data owners can require result auditing, so that they can see and approve (or reject) results generated by analysis of their datasets before these results become available to the data consumer.

• Processing results are registered on the blockchain and assigned a DOI compatible RAI Persistent Identifier (example https://doi.org/10.82136/212)..)

• When the RAI Persistent Identifier is cited in a paper, the data owner is also included in the citation’s metadata. 

• Data owners have as metrics how many times, when and by whom their datasets have been processed.

• Researchers can choose whether scripts and results produced by them are downloadable by others.

• The real dataset is never downloaded or visible by other researchers

RAISE is not just another data portal. RAISE is the infrastructure as a service for providing access to data.

• https://portal.raise-science.eu

• https://documentation.raise-science.eu

TRL: 7