ESA Open Invitation to Tender AO10358
Open Date: 06/08/2020
Closing Date: 19/10/2020 13:00:00

Status: ISSUED
Reference Nr.: 20.187.09
Prog. Ref.: Technology Developme
Budget Ref.: E/0901-01 – Technology Developme
Special Prov.: AT+BE+CH+CZ+DE+DK+EE+ES+FR+FI+GB+GR+HU+IT+IE+LU+NO+NL+PL+PT+RO+SE
Tender Type: C
Price Range: 200-500 KEURO
Products: Ground Segment / Mission Operations / Other
Technology Domains: Space System Software / Earth Observation Payload Data Exploitation / Data and Information Processing and Exploitation
Establishment: ESRIN
Directorate: Directorate of EO Programmes
Department: EO Mission Management & GS Department
Division: ESA Ground Segment & Data Mngmt Division
Contract Officer: Frigot, Emma
Industrial Policy Measure: N/A – Not apply
Last Update Date: 06/08/2020
Update Reason: Tender issue

Bulk Processing via Parallel computing (BULPP) is a Technology Development Element activity that will respond to the need to perform large data reprocessing or reformatting to apply refined or evolved algorithms and data formats, and to improve data exploitation considering advanced techniques suitable for handling very large data quantities in limited time, e.g., by making use of the same auxiliary data for all the scenes or benefiting from a continuous information flow of the satellite data take. A proven possibility for improving the processing speed is to take advantage of the massive parallelization capabilities of Graphical Processing Units (GPUs). This parallelization is achieved with hundreds or thousands of processing cores that are able to work concurrently.As such the expected output of BULPP should be a modular bulk processing framework for EO payload data processing, based on a parallel (GPU) computing approach considering the overall end-to-end process, from the input data repository, often containing the lowest level data together with the auxiliary data needed for the processing, the scientific algorithms used in the current implementation, up to the output repository which will include reprocessed data and their metadata.The bulk processing framework shall be applicable to a variety of EO satellite missions and products, benefitting from commonalities, frequent usage, scalability and portability onto current (and future) state-of-the-art parallel processing technologies. This usually implies not only a change in the physical architecture,but also a full re-thinking, re-designing and re-writing of the software code.The envisaged bulk processing framework shall aim atextending the portfolio of parallelised (and parallelisable) algorithms, increasing the portability across different HW infrastructures, allowing data intensive processing close to the data storage, thus reducing the demands on network bandwidth and data transfer.