> Through a foundational change, Amazon EKS has offloaded etcd’s consensus backend from a raft-based implementation to journal, an internal component we’ve been building at AWS for more than a decade. It serves ultra-fast, ordered data replication with multi-Availability Zone (AZ) durability and high availability.
I'd love to know more about this. Is it Paxos or something else?
> Through a foundational change, Amazon EKS has offloaded etcd’s consensus backend from a raft-based implementation to journal, an internal component we’ve been building at AWS for more than a decade. It serves ultra-fast, ordered data replication with multi-Availability Zone (AZ) durability and high availability.
I'd love to know more about this. Is it Paxos or something else?