I have some questions regarding snapshots to s3 and Intelligent Tiering, we are seeking to save some $$ on our snapshot storage, however, If the snapshot process needs to access past snapshots to create the new snapshots, Intelligent Tiering may not save us much at all. Has anyone used this storage class with snapshots? I know glacier is not supported by the current plugin, but ideally, it would be nice to use glacier in the future.
Thanks.
The snapshot process only really looks at a few metadata objects when performing subsequent snapshots, and doesn't look at most of the data that's there. I think it does list all the objects in the repository, but it's not clear from the AWS docs whether this Intelligent Tiering counts listing objects as "access". I'd hope not, and it seems worth a go.
I opened https://github.com/elastic/elasticsearch/issues/38836 recommending that we natively support this storage class in a future version. A PR is, of course, more than welcome 