Elasticsearch stateless roadmap

Reading this recently published blog post I wonder how was this indexing improvement achieved. I have two main questions:

  • What kind of storage was used for the stateful cluster? Was it network based storage or local SSD disks?
  • Were consistency guarantees reduced? If only segments are being replicated to the object store is there a point in time during which an indexing node holds the single copy of a segment or some segments?

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.