LAGO is a cosmic ray observatory composed of a network of water-Cherenkov detectors (WCD) spanning over different altitudes and latitudes making research on High Energy Physics, Space weather, Life Sciences, Aerospatial security, Computer Science, etc.
The measurements collected from these detectors are posteriorly processed and analysed. Additionally, scientists continuously generate simulated data. The final purpose is to enable the long-term curation and re-use of data within and outside LAGO through a Virtual Observatory.
The architecture of the thematic service is shown in figure. LAGO’s thematic service is focused on providing a standardised way to curate and reuse measurements, analysis and simulations. To achieve this task, it follows the basic design recommended by EGI/EOSC for cloud: core intelligence packed in Docker images, being able to automatically check, store and publish their results in DataHub, with enough metadata to be used by official harvesters (B2FIND), which will act as virtual observatories.
As the whole computation is self-contained in the image, the production can be easily performed by services such as EC2/IM or even manually in private clusters.
LAGO thematic service includes or will include the following services listed in the EOSC marketplace:
- EGI Check-in (through EduTeams Perun at GEANT): it is needed for accessing any EOSC service, in particular for obtaining a OneData token. Managing the VO with Perun at GEANT was considered because of flexibility and their long-term support to Latin American users.
- EGI DataHub: OneData allows researchers several ways to access the data and metadata of their interest. Collaboration members can directly explore the directory tree at https://datahub.egi.eu or mount it on their PC’s. Meanwhile, the general public will get published data through B2FIND. On the other hand, OneData eases storing results without modifying simulation/processing codes, as well as maintaining usable replicas around the world.
- The EOSC Cloud services (IM and EC3) will be explored in the coming months to validate the deployment of batch or Kubernetes clusters.
Additionally, LAGO plans to explore other services such as B2FIND, B2HANDLE and DIRAC4EGI.