Massively parallel Bayesian inference for transient gravitational-wave astronomy

Analysing transient gravitational-wave signals in the minimum possible wall time is crucial for identifying possilbe electromagnetic counterparts. In this work, we demonstrate how using massively parallel nested sampling can dramatically decrease the run time of the most expensive analyses. An implementation is available in the open source gravitational-wave inference library 𝚙𝙱𝚒𝚕𝚋𝚢 (parallel 𝚋𝚒𝚕𝚋𝚢).