The same hardware is still in use today, and I expect will continue to be used for some time.
The biggest thing that’s happened in the meantime is reducing load on the databases by pushing some of the OCSP load out to redis caches: https://letsencrypt.org/2022/12/15/ocspcaching.html By not storing OCSP in the MariaDBs, we reduced write volume and storage size significantly.
The next big thing is going to be sharding the database so multiple servers can share the write loads, and some queries moved out to redis caches.
We are beginning to hit the limits of what we can do with the current single-writer database in this hardware.
As well, the size of the data is larger than I’d like for a single MariaDB cluster. That makes operations like setting up new replicas take a long time. Being able to operate on smaller shards will help a lot at parallelizing some of that work.
The biggest thing that’s happened in the meantime is reducing load on the databases by pushing some of the OCSP load out to redis caches: https://letsencrypt.org/2022/12/15/ocspcaching.html By not storing OCSP in the MariaDBs, we reduced write volume and storage size significantly.
The next big thing is going to be sharding the database so multiple servers can share the write loads, and some queries moved out to redis caches.
(I work for Let’s Encrypt)