Derrick: A Three-Layer Balancer for Self-Managed Continuous Scalability

Data arrangement determines the capacity, resilience, and performance of a distributed storage system. A scalable self-managed system must place its data efficiently not only during stable operation but also after an expansion, planned downscaling, or device failures. In this article, we present Derrick, balancing algorithm addressing these needs, which has been developed for HYDRAstor, highly commercial Derrick makes decisions quickly in case failures takes additional time to find nearly optimal plan reaching it when population changes. Compared algorithms two other state-of-the-art systems, provides better capacity utilization, reduced movement, improved performance. Moreover, can be easily adapted meet custom placement requirements.