Navigation and service

Go to:

Federated Systems and Data

The division conducts research on methods and operates services for data management, data analytics, and the federation of resources such as data storages and HPC-systems. The investigation and development of concepts and tools for deep learning, the exploration of data for the earth system, and a direct, individual user support feed back into development and operations, thus foster the functionality and usability of all services, software, and infrastructures.
The division is the core development partner of the open source software UNICORE (Uniform interface to computing resources).

 

UNICORE Workflow created in the UNICORE Rich Client

Federation Services

The group Federation Services develops software for accessing compute and data resources and securely integrating them into federations. The main focus of our work is the UNICORE middleware, which includes adapters to many HPC resource managers like Slurm or LSF and provides access to a number of data sources such as POSIX file systems or Cloud storages. We continually update and extend our tools to help implement use cases such as scientific workflows, data sharing, community portals and many others.

More: Federation Services …

Distributed Systems and Infrastructures

Data Services and Infrastructure

The group Data Services and Infrastructure develops and operates generic data services for publication and sharing, as well as community specific data services. We also operate UNICORE and Cloud services integrated in various federated infrastructures.

More: Data Services and Infrastructure …

Earth System Data Exploration

Earth System Data Exploration

The research group Earth System Data Exploration (ESDE) develops innovative methods and tools for the integration and analysis of complex, heterogeneous, and big datasets related to air pollution, weather and climate. We work together with the Cross-Sectional Team Deep Learning and the Simulation Laboratories SL Climate Science and SL Terrestrial Systems to apply modern IT methods in order to advance the scientific analysis of earth system data from different observation systems and models. It benefits from FSD activities on federated data services.

More: Earth System Data Exploration …

Cross Sectional Team Deep Learning

Cross-Sectional Team Deep Learning

The Cross-Sectional Team Deep Learning conducts basic and applied research in the field of adaptive multi-layer neural network architectures, focusing on unsupervised and reinforcement learning of complex tasks from very large amount of raw, unprocessed data.

More: Cross-Sectional Team Deep Learning …

to the top

© Forschungszentrum Jülich Legal Notice Data protection