Ursa minor: versatile cluster-based storage

TLDR

No single encoding scheme or fault model is optimal for data, so a versatile storage system that matches them to access patterns, reliability requirements, and cost goals per data item is needed. The study proposes a versatile storage system that matches encoding schemes and fault models to access patterns, reliability requirements, and cost goals per data item. Ursa Minor is a cluster‑based storage system that enables data‑specific selection and online adjustment of encoding schemes and fault models. Experiments demonstrate that using specialized encoding and fault‑model choices in Ursa Minor yields 2–3× performance improvements, nearly doubles aggregate cluster throughput, and allows a single cluster to efficiently support multiple workloads compared to a one‑size‑fits‑all configuration.

Abstract

No single encoding scheme or fault model is optimal for data. A versatile storage system allows them to be matched to access patterns, reliability requirements, and cost goals on a per-data item basis. Ursa Minor is a cluster-based storage system that allows data-specific selection of, and on-line changes to, encoding schemes and fault models. Thus, different data types can share a scalable storage infrastructure and still enjoy specialized choices, rather than suffering from one size fits all. Experiments with Ursa Minor show performance benefits of 2-3× when using specialized choices as opposed to a single, more general, configuration. Experiments also show that a single cluster supporting multiple workloads simultaneously is much more efficient when the choices are specialized for each distribution rather than forced to use a one size fits all configuration. When using the specialized distributions, aggregate cluster throughput nearly doubled.