Deep Hybrid Data Cloud (DHDC)

  • contact:

    Dr. Marcus Hardt

  • funding:


  • startdate:


  • enddate:


The DHDC project investigates how to support compute-intensive applications that require high-performance computing (HPC) and graphics processors (GPUs) with the help of cloud services.

Deep Hybrid Data Cloud (DHDC)

In the Deep Hybrid Data Cloud (DHDC) project, ten partners from the successfully completed INDIGO DataCloud project have joined forces. Fifteen full-time employees will spend 30 months on cloud-based solutions that support complex, compute-intensive applications. The project budget is three million euros.

DHDC focuses on applications that require high-performance computers (HPC) and graphics processors (GPUs). A "hybrid-cloud approach" is intended to provide scientists with easy access to cloud resources. To this end, a series of building blocks will be developed under the "DEEP as a Service" label, which will provide the simple development and scaling of applications that meet these requirements. These applications include deep learning methods, neural networks, parallel processing of large amounts of data and the online analysis of streaming data. In addition, components of the predecessor project INDIGO DataCloud are to be integrated, which allow the user to configure the interaction of DHDC modules in a simple way. The finished building blocks will then be further developed into services that are integrated and run in a scaled manner within the framework of the European Open Science Cloud (EOSC).

Contact at SCC: Dr. Marcus Hardt