UNAVCO GNSS data operations modernization and uplift to the cloud: real-time and archiving.

UNAVCO operates the NSF repository for geodetic GNSS data. We are transitioning from on-premise data operations to a cloud-based solution over the next two years. As part of this uplift, we will modernize our data operations utilizing many of the Apache big data products such as Kafka and Flink. Our data flow will be containerized and run in Kubernetes. The real-time GNSS data streams will continue to be distributed with NTRIP casters, but the internal data path will be updated, made more robust, and be the primary way of collecting data, as opposed to downloading daily files. We also plan to move our archive from static daily files hosted on an ftp/http server to a microservice that will deliver the GNSS data in multiple formats and over user selected date ranges. The formats will include, but not be limited to RTCM, NetCDF, HDF5, RINEX 2, RINEX 3, and BINEX. The data can be filtered or downsample when downloading. Furthermore, there will be an option to receive the data packed as satellite arcs in addition to the traditional time slices with the internal data structure optimized to provide both data types. This service will be wrapped into an http server to emulate the currently daily file structure with microservice calls to support existing data discovery models while incorporating more modern discovery models as well.