A robust framework for multicloud management using machine learning and optimization techniques

Abstract

In recent years, enterprise adoption towards cloud is rapidly increasing and every company is keen on moving towards cloud for various business reasons like cost reduction, productivity improvement, operational efficiency and short time frame to market their solutions. Companies have started using either single or multiple clouds based on their investment and business needs. Multicloud and hybrid cloud management is gaining lots of momentum in large enterprises and in many organizations because of their business demand and to avoid vendor lock-in. Moreover, companies are keen on making their applications more portable, distributed and interoperable and want applications to be more collaborative in nature. Hence there is a shift in focus towards multicloud environment by organizations who intend to be more flexible and adopt to multicloud, to meet their business demands. Multi-cloud architecture enables organizations to distribute their workloads across different clouds so that they can best utilize their services and have big cost saving while mitigating risks associated with individual cloud environments. Multi-cloud environment offers a rich set of cloud services in which some are common and some are very unique. These services are designed to solve various needs across a diverse range of computing and business functions, thereby optimizing returns on cloud investments. Multi-cloud allows organizations to utilize best services of each cloud platforms in order to support the regional and country specific business needs, government policies and local customer demands. With this minimum vendor lock-in, customers get business benefits on performance, security and good returns on investments (ROI) while addressing changing business needs, and have flexibility to move to any cloud easily with diverse technology support. While there are increasing adoptions towards multicloud, there are many operational challenges in managing multicloud. Any poor planning in multicloud adoption can lead to challenges like interoperabi