de.KCD

The German Competence Center Cloud Technologies for Data Management and Processing (de.KCD) is a cross-location and cross-domain contact point for teaching skills in handling data using cloud-based technologies, resources and methods for institutions and networked centers as well as for researchers at all career levels.

In the current phase of digital transformation, scientific and economic success as well as the connectivity of future-oriented projects depend on (i) the systematic and structured collection of relevant raw and metadata through expertise-based data management, (ii) the development and provision of innovative cloud offerings and automated workflows, (iii) the development and transfer of expertise in the field of cloud-based data processing, and (iv) the availability of a powerful and independent cloud infrastructure.

The de.KCD specifically addresses these challenges in order to bundle expertise in cloud computing and data management, as well as to provide and expand the necessary hardware capacities and cloud services. The project implements suitable measures for cloud-based data management and standardized data analysis. It offers cloud infrastructure, storage and analysis options as well as generic training for knowledge transfer across different specialist areas. Virtual learning and working environments lower the barriers to access. In addition, technologies are being developed to answer complex research questions and integrate data from different disciplines. Beyond these measures, de.KCD also aims to promote collaboration and knowledge exchange between research locations by creating a networked, collaborative data space for national and international research projects.

Kernkompetenzen und Services

Cloud-based Data Management

With regard to specific data science methods, we primarily want to teach the use of cloud-based infrastructures for distributed and scalable data management as well as the necessary skills for standardized and automated data processing. In accordance with the FAIR principles, this includes specific expertise for the reproducible handling of data and the use of appropriate software tools. This is made possible, for example, through the use of software container solutions (e.g. BioContainers) in conjunction with cloud-based data management systems (e.g. distributed databases).

Workflows

For scalable and automated processing of research data, we teach skills in workflows (e.g. Nextflow or Galaxy) that ensure easier traceability (data provenance) of the collection, generation, processing and reproducibility of research data (e.g. as machine-readable results in the form of Research Data Objects). Knowledge of the evaluation and use of version control systems for the audit-proof storage of e.g. differently parameterized workflows or individual evaluation scripts completes the data competence transfer in this area.

Software Stacks

Flexible virtualization and the use of different and specialized software solutions (e.g. GPU-based algorithms) for cloud-based data analysis often requires the use of dedicated cloud-based software stacks. To this end, scientists and data analysts from all specialist areas are actively supported in the development and establishment of tailored software stacks, which are periodically renewed and automatically tested in order to automate and scale the underlying data management and the management of the virtualized compute environment in the best possible way.

Trusted Research Environments

In order to enable secure and data protection-compliant processing of particularly sensitive or personal data in the cloud, requirements for Trusted Research Environments (TREs) will be described and material for the automated setup of these in cloud environments will be designed and tested with the aim of making them available at certified cloud locations and also using them for training. As different technical and organizational implementation models are possible, these will be presented in dedicated user meetings involving users and existing providers of TREs and compared in terms of their professional, legal and technical advantages and disadvantages.

Wissens- und Kompetenzbasis

In order to train researchers to perform their own data processing and analysis using various cloud technologies, we are developing a structured training program and self-learning units consisting of learning paths and modules, complemented by a scalable, cloud-based training infrastructure with preconfigured learning environments in de.KCD. As a whole, this forms the knowledge and competence base of de.KCD and represents a community-driven framework for the collection of FAIR training materials for software developers, system administrators and scientists. We place particular emphasis on train-the-trainer lessons so that the generic materials can be transferred to other subject-specific domains and used there, in an adapted form, for corresponding training courses.

Online and Face-to-Face Courses

We offer our training content in the form of online, hybrid and face-to-face courses throughout the year for researchers at all career levels, which are announced in our basic portal with an integrated training calendar. These formats will be complemented by annual summer schools to provide participants with an intensive and interactive learning experience where they can acquire new knowledge, deepen their understanding of a particular subject area and engage with other participants and experts on topical issues. The summer schools will also promote networking among participants, thus strengthening collaboration and exchange in the academic community.