AMIGA : Participation in the SKA Science Data Processor consortium

The main challenge of Radiostronomy for the next decade will be to build the Square Kilometre Array (SKA). This instrument, with the potential to pro

The main challenge of Radiostronomy for the next decade will be to build the Square Kilometre Array (SKA). This instrument, with the potential to provide revolutionary science, will address core questions in the formation and evolution of galaxies: the buildup of stellar mass and the role of neutral gas, in and around galaxies, key for star formation (SF). While cosmological cold gas accretion is suspected to maintain SF, HI stripping might inhibit it in dense environments. Both extremes of the HI lifecycle, still poorly understood since their low column densities are only reachable by the SKA, are currently the target of AMIGA group. AMIGA science focuses on HI observations of galaxies, with complex 3D data products in a typical 12-h observing run of SKA1 reaching 300GB, involving data rates of 102 GB/s. These numbers will transform the way scientists will access and analyse observations. For this reason AMIGA preparatory work for the SKA will complement fundamental science with applied research. The SKA Science Data Processor consortium is led by the University of Cambridge and is in charge of designing the system to transform the data coming from the correlator in to science ready data products and deliver them to the astronomers. As a part of this consortium, AMIGA's main goalis to reinforce the efficient distribution of the SKA data and metadata among the scientific community. Specifically, it targets to: Provide the astronomers with a transparent access to the data and to exploit the computing resources minimizing the latency time and the energy consumption. Characterize SDP data and processes. These should go along with a metadata layer during their whole life cycle and final data products should include the key parameters that were used during their creation and preprocessing. Evaluate the suitability of different hardware configurations. AMIGA members participate in three SDP working package: DATA, DELIV and PROT. DATA includes the definition of the data model for the SKA Archive, the identification of data flow and data life-cycle requirements. DELIV covers the delivery of data to the end users, researching solutions for data discovery, visualisation and transference to the Regional Centers (RCs), as well as the design of the interface to the RCs. PROT is in charge of designing different prototypes in order to test and asses the validity of different parts of the whole SDP. These activities is funded under grantAYA2015-65973-C3-1-R(MINECO/FEDER, UE),and is a collaboration between astronomers atIAA(CSIC) and theFCSCL supercomputing centre, the Universidad de Granada (UGR), the Universidad Politécnica de Madrid (UPM), the Universidad Politécnica de Valencia (UPV) and the Real Insituto y Observatorio de la Armada de San Fernando (ROA) as well as the Barcelona Supercomputing Center (BSC).
 

Participation in the SKA Science Data Processor consortium

The main challenge of Radiostronomy for the next decade will be to build the Square Kilometre Array (SKA). This instrument, with the potential to provide revolutionary science, will address core questions in the formation and evolution of galaxies: the buildup of stellar mass and the role of neutral gas, in and around galaxies, key for star formation (SF). While cosmological cold gas accretion is suspected to maintain SF, HI stripping might inhibit it in dense environments. Both extremes of the HI lifecycle, still poorly understood since their low column densities are only reachable by the SKA, are currently the target of AMIGA group.

AMIGA science focuses on HI observations of galaxies, with complex 3D data products in a typical 12-h observing run of SKA1 reaching 300GB, involving data rates of 102 GB/s. These numbers will transform the way scientists will access and analyse observations.  For this reason AMIGA preparatory work for the SKA will complement fundamental science with applied research.

The SKA Science Data Processor consortium is led by the University of Cambridge and is in charge of designing the system to transform the data coming from the correlator in to science ready data products and deliver them to the astronomers. As a part of this consortium, AMIGA's main goal is to reinforce the efficient distribution of the SKA data and metadata among the scientific community. Specifically, it targets to:

  • Provide the astronomers with a transparent access to the data and to exploit the computing resources minimizing the latency time and the energy consumption.
  • Characterize SDP data and processes. These should go along with a metadata layer during their whole life cycle and final data products should include the key parameters that were used during their creation and preprocessing.
  • Evaluate the suitability of different hardware configurations.


AMIGA members participate in three SDP working package: DATA, DELIV and PROT. DATA includes the definition of the data model for the SKA Archive, the identification of data flow and data life-cycle requirements. DELIV covers the delivery of data to the end users, researching solutions for data discovery, visualisation and transference to the Regional Centers (RCs), as well as the design of the interface to the RCs. PROT is in charge of designing different prototypes in order to test and asses the validity of different parts of the whole SDP.

These activities is funded under grant AYA2015-65973-C3-1-R (MINECO/FEDER, UE), and is a collaboration between astronomers at IAA (CSIC) and the FCSCL supercomputing centre, the Universidad de Granada (UGR), the Universidad Politécnica de Madrid (UPM), the Universidad Politécnica de Valencia (UPV) and the Real Insituto y Observatorio de la Armada de San Fernando (ROA) as well as the Barcelona Supercomputing Center (BSC).