6/2024 - Project explanation and research updates of the DAPHNE project
The DAPHNE project aims to define and build an open and extensible system infrastructure for integrated data analysis pipelines, including data management and processing, high-performance computing (HPC), and machine learning (ML) training and scoring. The acronym DAPHNE relates to the title "integrated Data Analysis Pipelines for large-scale data management, High-performance computing, and machiNE learning".
In DAPHNE, the team of the DLR Institute of Data Science is investigating operations
in main memory, storage class memory (SCM), secondary storage (e.g., SSDs), and tertiary storage
(e.g., tape). Modern storage devices for transient and persistent storage provide means of near-data
processing with similar abstractions, which allows avoiding unnecessary data movement along the storage and memory hierarchies in order to improve performance and reduce the energy consumption of data movement.
With a large consortium of international partners, including Know-Center GmbH, ETH Zurich, TU Dresden, Intel Technology Poland, IT University of Copenhagen in direct collaboration, DAPHNE is the cornerstone for an open source community for the development of an open infrastructure for integrated data analysis pipelines. The results of the project are available via Github in regular releases (https://github.com/daphne-eu).
Read more about the DAPHNE project on the project homepage or here.