Gravitational wave alert generation infrastructure on your laptop

Multi-messenger astrophysics provides valuable insights into the properties of the physical Universe. These insights arise from the complementary information carried by photons, gravitational waves, neutrinos and cosmic rays about individual cosmic sources and source populations. When a gravitational wave (GW) candidate is identified by the Ligo, Virgo and Kagra (LVK) observatory network, an alert is sent to astronomers in order to search for electromagnetic or neutrino counterparts. The current LVK framework for alert generation consists of the Gravitational-Wave Candidate Event Database (GraceDB), which provides a centralized location for aggregating and retrieving information about candidate GW events, the SCiMMA Hopskotch server (a publishsubscribe messaging system) and GWCelery (a package for annotating and orchestrating alerts). The first two services are deployed in the Cloud (Amazon Web Services), while the latter runs on dedicated physical resources. In this work, we propose a deployment strategy for the alert generation framework as a whole, based on Kubernetes. We present a set of tools (in the form of Helm charts, Python packages and scripts) which conveniently allows running a parallel deployment of the complete infrastructure in a private Cloud for scientific computing (the Cloud at CNAF, INFN Tier-1 Computing Centre), which is currently used for integration tests. As an outcome of this work, we deliver to the community a specific configuration option for a sandboxed deployment on Minikube, which can be used to test the integration of other components (i.e. the lowlatency pipelines for the detection of the GW candidate) with the alert generation infrastructure in an isolated local environment.