The IceCube Neutrino Observatory consists of two detector components: a cubic-kilometer in-ice neutrino detector and the IceTop surface array for measuring extensive air showers generated by high-energy cosmic rays. The proposed next-generation neutrino observatory, IceCube-Gen2, will increase the in-ice instrumented volume and add an additional in-ice radio detector for high-energy neutrinos. Furthermore, it will feature an entirely new surface array comprising approximately $160$ surface stations covering more than $6\,$km$^2$, each equipped with eight scintillator panels for charged particle detection and three antennas for measuring air-shower radio emission.
The surface radio readout of each station will receive triggers not only from local coincidences of the scintillators, but also from the reconstruction of other IceCube and IceCube-Gen2 detectors, requiring sufficient buffer depth of several seconds.
To achieve this buffer time, while aiming at $40\,$W electrical power budget per station, an Adaptive Readout for Gigasample Operation (ARGO) board is developed, which uses a multi-FPGA approach with LPDDR5 memory for radio buffering. Radio front-end and scintillator readout are realized with optional daughterboards to provide an experiment-agnostic mainboard that could also be used in other air-shower experiments.
In this contribution we will present the concept of the ARGO board and its current development status.