Architectural partitioning and deployment modeling on hybrid clouds

Abstract

The hybrid cloud idea is increasingly gaining momentum because it brings distinct advantages as a hosting platform for complex software systems. However, there are several challenges that need to be surmounted before hybrid hosting can become pervasive and penetrative. One main problem is to architecturally partition workloads across permutations of feasible cloud and non-cloud deployment choices to yield the best-fit hosting combination. Another is to predict the effort estimate to deliver such an advantageous hybrid deployment. In this paper, we describe a heuristic solution to address the said obstacles and converge on the ideal hybrid cloud deployment architecture, based on properties and characteristics of workloads that are sought to be hosted. We next propose a model to represent such a hybrid cloud deployment and demonstrate a method to estimate the effort required to implement and sustain that deployment. We also validate our model through dozens of case studies spanning several industry verticals and record results pertaining to how the industrial grouping of a software system can impact the aforementioned hybrid deployment model.As alluded to earlier, hybrid environments are composed of multiple cloud and traditional data centers; the cloud is typically multi-vendor and heterogeneous, comprising private, public, and community ('shared private') deployments. While the deployment of future IT systems will turn increasingly hybrid [7], the challenge is to arrive at a topology that results in the least cost of ownership. There are many permutations of hybrid deployment architectures available that offer varying degrees of control, isolation, security, and performance. One or more of these permutations is likely to offer the most advantageous fitment depending on the characteristics of various categories of component workloads.The problem, thus, is to design a deployment strategy for the service provider that apportions different workloads of a business system across a hybrid infrastructure that satisfies various nonfunctional requirements related to performance, security, and availability at the minimum cost of ownership. A cloud deployment, at a high level, comprises the cloud stack (which we refer to as the 'managing environment') and the application workloads that run on target virtual machines ARCHITECTURAL PARTITIONING AND DEPLOYMENT MODELING ON HYBRID CLOUDS 347 (which we call the 'managed environment'). In this paper, we focus on the complexity and effort in constructing the managing environment.An additional problem is to model the effort required to implement such a hybrid cloud-based deployment, which can then be used as a prediction tool for techno-commercial decisions spanning architecture and business. 1.2C5C1C2/ or 0.1 as the number of workloads in Quadrant #3 increase beyond the threshold dictated by ı w . Similarly, the constant in the denominator of the fourth term ‡ Although we focus on Q1 here, we also repeated this experiment for other quadrants and obtained similar results. We did not ...

Building a blockchain-based decentralized ecosystem for cloud and edge computing: an ALLSTAR approach and empirical study

“…On the other hand, academic works mainly focus on Cloud resource allocation and scheduling. Venkateswaran and Sarkar (2018) optimize the application deployment process on hybrid Clouds and achieve the best-fit hosting combination. Wang et al (2011) enhance the makespan and the reliability of a workflow application through evaluating the Cloud resource reputation.…”

ALLSTAR: A Blockchain Based Decentralized Ecosystem for Cloud and Edge Computing

“…On the other hand, academic works mainly focus on Cloud resource allocation and scheduling. Sreekrishnan et al [4] optimize the application deployment process on hybrid Clouds and achieve the best-fit hosting combination. Xiaofeng et al [5] enhance the makespan and the reliability of a workflow application through evaluating the Cloud resource reputation.…”

CloudsStorm: A framework for seamlessly programming and controlling virtual infrastructure functions during the DevOps lifecycle of cloud applications

“…In the academic area of cloud computing, there has been plenty of papers working on how to control the infrastructure to satisfy the application requirements. For example, Venkateswaran and Sarkar proposed a model to deploy applications on hybrid clouds and yield the best‐fit hosting combination. Ziafat and Babamir focused on the algorithm selecting a proper cloud to run the task according to the geographical location of the data center.…”