A Taxonomy and Survey of Cloud Resource Orchestration Techniques

Abstract

Cloud services and applications prove indispensable amid today’s modern utility-based computing. The cloud has displayed a disruptive and growing impact on everyday computing tasks. However, facilitating the orchestration of cloud resources to build such cloud services and applications is yet to unleash its entire magnitude of power. Accordingly, it is paramount to devise a unified and comprehensive analysis framework to accelerate fundamental understanding of cloud resource orchestration in terms of concepts, paradigms, languages, models, and tools. This framework is essential to empower effective research, comprehension, comparison, and selection of cloud resource orchestration models, languages, platforms, and tools. This article provides such a comprehensive framework while analyzing the relevant state of the art in cloud resource orchestration from a novel and holistic viewpoint.

Towards Secure Cloud Orchestration for Multi-Cloud Deployments

“…Weerasiri et al [24] lists security as one of the crosscutting concerns in cloud orchestration implementations, along with service level agreements and negotiations, portability, interoperability, standardization, resource demand profiling, resource pricing, profit maximizing and other runtime issues. The study, found that such cross-cutting concerns are addressed both by research initiatives, and to a larger extent by proprietary orchestration products, such as: Amazon Web Service OpsWorks, CFEngine, Docker, Heroku, Juju, Puppet, and VMWare vSphere.…”

“…The set of high-level attacks is based on earlier research [18,20] and review articles [3,7,22,24]. VM Substitution Attack The adversary induces the orchestrator to launch a VM instance that contains hidden vulnerabilities, using a maliciously modified VM image instead of the tenant-selected image.…”

QoS-Aware Orchestration of Network Intensive Software Utilities within Software Defined Data Centres

“…The work of Weerasiri et al [48] presents a taxonomy and survey of cloud resource orchestration techniques, which includes autonomic orchestration strategies.…”

“…Existing orchestration systems [27,19,48] have been designed mainly for single data centres to help improve the make-span of workflow-based applications. QoS aspects related to networking have been traditionally addressed with Software Defined Networking (SDN) approaches [12], such as the one of Wang et al [47].…”

On-demand virtual research environments using microservices

“…Nevertheless, this way of procuring infrastructure introduces a layer of complexity that researchers may find hard to cope with. Configuring virtual resources requires substantial technical skills 5 , and it is generally a tedious and repetitive task when infrastructure is allocated on demand. Therefore, when running scientific applications on cloud there is a need for a methodology to aid this process.…”