Software-defined storage (SDS) is a new generation of storage systems. Unlike the traditional embedded-system based storages, the SDS uses a software-stack above commodity hardware to provide more valuable and cost-effective features. This paper argues that the architecture of an SDS platform should have three main characteristics: 1. The separation of the control and data path, 2. Self-configuration of storage resources, and 3. RESTful APIs for new business extension. In this paper, the architecture of an SDS platform, called Federator, is first discussed. A new approach designed within the Federator for a self-configured SDS is described. This approach includes two types of neural networks, which provide optimal storage resource configuration for any type of application. It also includes fuzzy inference, which transforms the resultant optimal configuration into rules for automatic management.
SDS has become one of the most important concepts in the storage industry, not only because it is a cost-effective alternative to traditional high-cost storage systems, but also its ease-of-use and the flexibility it can engender in an ever-changing business environment. With the clear separation of the control and the data path, the intelligent self-configuration technologies, and the standard RESTful API, the Federator is expected to better meet requirements of new applications in ever-changing computing environments.