Failures In Distributed Databases

The “Morning Paper” blog has some fascinating insights into critical production failures in distributed data-intensive databases such as Cassandra and HBase. This reveals that simple testing can prevent most critical failures. In this blog post we take a quick look to see what an embeddable Paxos algorithm project such as Trex can learn from this study.  Read the rest of this entry »