Learn how to leverage external persistent volumes to enhance recovery times and achieve high availability on DC/OS in this informative conference talk. Explore the challenges faced by modern distributed applications like Cassandra and HDFS when dealing with node failures and data replication across failure zones. Discover how the recovery process to pre-failure redundancy levels can be time-consuming due to large data transfers to new nodes. Examine the impact on write operations during node bootstrapping and its effect on overall recovery time. Gain insights from Dinesh Israni, a Senior Software Engineer at Portworx with extensive experience in distributed storage solutions, as he shares strategies to overcome these challenges and improve system resilience in DC/OS environments.