Development and research of models of organization storages based on the software-defined infrastructure

Abstract

Data centers are widely used for the placement of highly loaded applications and solutions used for the processing of large data sets (BigData). For effective access to services, minimal response time of data storages is required. We explored a new paradigm in software-defined infrastructure distributed cloud systems. Also the design principles and high-level architectural design of the SD infrastructure controller are presented here. Modern applications and services needs constantly transforming infrastructure for ensuring consumer requirements (SLAs) amidst provider constraints (costs). This paper examines infrastructure in the context of platform for new applications and services and discussing some of the fundamental characteristics required of such infrastructure called software-defined infrastructure (SDI). We developed the models of the software-defined infrastructure. We performed an experiment to analyze the productivity of software-defined storage. Our experiment has shown that software-defined storage and scheduling algorithm in software-defined infrastructure placement can gain growth performance compared with the physical storage and virtual machines. This is necessary when storage systems work with high intensities requests.