It looks like they use something very similar to Erasure Coding in Ceph with RadosGW [1].
[1] https://ceph.com/planet/using-erasure-coding-with-radosgw/
[1] https://ceph.com/planet/using-erasure-coding-with-radosgw/
This article is about single-region storage vs. multi-region storage (and how to reduce the cost in this case). There is very little public info available about distributed storage systems in multi-region setup with significant latency between the sites.