Aggregation of cloud providers: A review of opportunities and challenges

Abstract

A single cloud provider alone is not able to satisfy the increasing demand for its services as the number of online users and services deployed by organizations grow exponentially. To achieve the promises of cloud computing, the necessity of aggregating different cloud provider into a single framework emerges. This concept overcomes the limitations of interoperability, portability, vendor lock in problem, scalability and elasticity of contemporary cloud computing. Collaborating with different cloud providers is not an easy task because each cloud provider has different kinds of architecture, virtualization technology, service level agreement, enterprise solution and interface for accessing services. This scenario is identified by many synonyms in the literature such as cloud federation, Intercloud, Multi-Cloud etc. This paper discusses various driving forces and advantages that attract researchers towards it. Further, this paper concentrates on exploring related work done in addressing different issues and finding gaps that will create opportunities for further research and development. This work will tend to emerge as a step towards the next-generation enterprise solution.

Understanding the challenges and novel architectural models of multi-cloud native applications – a systematic literature review

“…In particular, we present an overview of the body of knowledge found as a result of this review (considering the 88 finally selected primary studies). The primary studies were classified according to the research method used in the study, as defined in Hybrid Cloud (public vs. private) [44][45][46][47][48][49][50][51][52] IoT or physical nodes [53][54][55][56] Different cloud services from different cloud providers [8,17,26,29,31,34,[57][58][59][60][61][62][63][64][65][66][67][68][69][70][71][72][73] Multi-cloud in a broader sense [74][75][76] Federated Cloud Services [17,32,[77][78][79][80][81] Multi-cloud Services Communities [82] opinion pap...…”

“…In this primary study the authors referred to the physical distance between the client and the Cloud resources to classify them under private if the cloud vendor is located far away from client, or public, if the cloud vendor and the client are in nearby premises. In this respect 8,5% of the studies [17,32,[77][78][79][80][81], referred to federated cloud services, i.e. different cloud services coming from different or same providers, but that are already interoperable in the sense that an intermediate layer already provides the necessary capabilities for the usage of these services despite their peculiarities given that they are provided by different cloud providers.…”

“…Others, such as security, may need to be considered from another point of view, as a consolidated vulnerability model as point of reference for the concrete case of multi-cloud scenarios is lacking [101]. Therefore, different methods need to be explored for an efficient utilization of available resources [79]. Another consequence from the cloud service heterogeneity and the distributed nature of multi-cloud applications is that code must be partitioned and distributed between different cloud services.…”

“…The heterogeneity of cloud resources is driving the challenges in the deployment phase too. The analysis and comparation of various cloud solutions is still a hard task, with several open issues such as the lack of standards and models for cloud services description and operation [33,104] or the mapping of subsets of services/ components to physical resources in an optimized manner [48,69,79,81,101].…”

Recent advancements in resource allocation techniques for cloud computing environment: a systematic review

“…For these purposes, predicted resource allocation is applied to allocate or reserve the resources for the future before they are needed [56]. It is significant and essential for effective resource allocation in IaaS cloud computing [57].…”