Modern cloud databases are shifting from converged architectures to storage disaggregation, enabling independent scaling and billing of compute and storage. However, cloud databases still rely on external, converged coordination services (e.g., ZooKeeper) for their control planes. These services are effectively lightweight databases optimized for low-volume metadata. As the control plane scales in the cloud, this approach faces similar limitations as converged databases did before storage disaggregation: scalability bottlenecks, low cost efficiency, and increased operational burden. We propose to disaggregate the cluster coordination to achieve the same benefits that storage disaggregation brought to modern cloud DBMSs. We present Marlin, a cloud-native coordination mechanism that fully embraces storage disaggregation. Marlin eliminates the need for external coordination services by consolidating coordination functionality into the existing cloud-native database it manages. To achieve failover without an external coordination service, Marlin allows cross-node modifications on coordination states. To ensure data consistency, Marlin employs transactions to manage both coordination and application states and introduces MarlinCommit, an optimized commit protocol that ensures strong transactional guarantees even under cross-node modifications. Our evaluations demonstrate that Marlin improves cost efficiency by up to 4.4x and reduces reconfiguration duration by up to 4.9x compared to converged coordination solutions.

We present BORA (Bag Optimizer for Robotic Analysis), a file system middleware that optimizes the acquisition of bags, which are specially formatted files used to store timestamped ROS (robot operating system) messages. BORA sits between ROS and an existing file system to conduct semantic-aware data pre-processing. In particular, it categorizes ROS bag data into multiple groups with each having a distinct label. BORA predigests data index constructions and reduces file open time via a hash-based label management scheme. It is also capable of providing ROS analytic applications with only data needed without a sequence of data searching and locating operations. We implement a BORA prototype, which is then integrated into three computing platforms: a single-node server, a four-node PVFS storage cluster, and a Tianhe-1A Supercomputer storage subsystem. Next, we evaluate the BORA prototype on the three platforms using four real-world ROS applications. Our experimental results show that compared to a traditional bag management scheme BORA improves data acquisition performance by up to 11x. In addition, it offers up to 10x data acquisition performance improvement and 3,100x bags open improvement under a swarm robotics data analysis scenario where data is retrieved across multiple bags simultaneously.