October 5-9, 2014

Abstract

P2.5 Framework for the integration of multi-instrument pipelines in the BepiColombo Science Operations Control System

Fernando Perez-Lopez (ESA/ESAC/GMV)

Juan Carlos Vallejo, Santa Martínez, I, Ortiz de Landaluce, Alan Macfarlane, Pedro Osuna, Ranpal Gill, Mauro Casale

BepiColombo is an interdisciplinary ESA mission to explore the planet Mercury in cooperation with JAXA. The mission consists of two separate orbiters: ESA’s Mercury Planetary Orbiter (MPO) and JAXA’s Mercury Magnetospheric Orbiter (MMO), which are dedicated to the detailed study of the planet and its magnetosphere. The MPO scientific payload comprises 11 instruments covering different scientific disciplines developed by several European teams. The Science Ground Segment (SGS), located at the European Space Astronomy Centre (ESAC) in Madrid, Spain, will be in charge of preparing the science operations for MPO. The SGS functionalities are: scientific payload operations planning, science operations, data processing and distribution to the instrument teams, preliminary analysis of the scientific data and scientific data archiving in a central archive accessible to the science community. The approach selected in BepiColombo MPO for the processing, analysis and archiving of the science data represents a significant change with respect to previous ESA planetary missions. Although traditionally, Instrument Teams are responsible for processing, analysing and preparing their science data for the long-term archive, in BepiColombo, the SGS will play a key role in these activities. The SGS has the full responsibility for the production of the first level of science data (un-calibrated), with the associated operational implications. In addition, the SGS will co-develop in collaboration with some of the instrument teams, part of their pipeline software and, in other cases, will be only responsible for the software integration. All aspects related to the integration of the software such as: different required pipeline execution platforms, pipeline software languages, pipeline development approach (externally developed or co-developed), different instrument pipeline development schedules have been considered. An additional difficulty that will impact the processing strategies is related to the various spacecraft data downlink mechanisms available for the MPO and their associated data latency. This paper describes the design and early implementation of the framework that copes with distributed data processing pipelines, possibly residing in different locations and developed by different teams, but designed as a single entity. Operationally, a primary-redundant pipeline configuration has been selected. Some instrument pipelines will be operated from the instrument team’s data processing centres, having a pipeline replica that can be run from the SGS, while others will be executed as primary pipelines from the SGS, with the SGS adopting the pipeline orchestration role. This role is based on a set of basic principles. Firstly, a common approach for pipeline data feeding and products collection is implemented by a generic component that will take the latest versions of the input data from the BepiColombo Archive Core System (BACS) providing them to the instrument pipelines (local or remote) and will get the products generated executing basic checks to ensure that they cover the interface requirements. Secondly, an innovative hierarchical classification of pipeline processes based on: tasks (which are the basic execution units), chains (where the tasks execution flow is specified), and recipes (a sequence of chains); which allow the configuration of the different execution flows and facilitates the reuse of common software modules between different pipelines. Finally, a monitoring and control functionally that triggers the execution (the cooking) of the instrument pipeline recipes and monitors their progress. The paper also describes the system engineering methodology, based on modularity and flexibility, applied and the integration of this framework within the whole BepiColombo Science Operations Control System

Mode of presentation: poster

Applicable ADASS XXIV theme category: Data Analysis / Pipelines