On Metastable Failures and Interactions Between Systems

Pile of Eternal Rejections: The Cost of Garbage Collection for State Machine Replication – Aleksey Charapko

Viewstamped Replication Revisited

Paxos vs. Raft: Have we reached consensus on distributed consensus?

Scalable but Wasteful, or why fast replication protocols are slow

Metastable Failures in Distributed Systems

Reading Group. A large scale analysis of hundreds of in-memory cache clusters at Twitter. | Aleksey Charapko