A mixed integer linear programming optimization approach for multi-cloud capacity allocation

Ciavotta, Michele; Ardagna, Danilo; Gibilisco, Giovanni Paolo

doi:10.1016/j.jss.2016.10.001

Cited by 15 publications

(14 citation statements)

References 40 publications

(46 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two of our previous works [17], [63] deal with problems similar to the one addressed in this paper. [17] introduces the MILP model exploited by the Initial Solution Builder (presented Section 4) to provide an initial solution to the problem.…”

Section: Related Workmentioning

confidence: 98%

“…In addition, that work assesses the impact of the adoption of this initial solution on the performance of SPACE4Cloud and the quality of the final solution. In [63] we generalize the model presented in [17] for the scenario of the capacity allocation problem of an application operated across multiple clouds simultaneously. The proposed MILP model identifies an initial deployment consisting in, for each hour of the day and each cloud provider, the type and number of virtual machines to be allocated to each application tier as well as the fraction of the total workload to be routed to the particular cloud.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Architectural Design of Cloud Applications: A Performance-Aware Cost Minimization Approach

Ciavotta¹,

Gibilisco²,

Ardagna³

et al. 2022

IEEE Trans. Cloud Comput.

Self Cite

View full text Add to dashboard Cite

Cloud Computing has assumed a relevant role in the ICT, profoundly influencing the life-cycle of modern applications in the manner they are designed, developed, and operated. The Cloud market offers highly diversified services upon which developers and operators can rely. Yet, its full adoption requires specific and rare expertise. Actually, such services are characterized by a steep learning curve and, among the other aspects, by significantly different Quality of Service (QoS). In this paper, we tackle the problem of supporting the design-time performance analysis of Cloud applications and the identification of the optimal strategy for allocating components onto Cloud resources. In particular, the final goal is to set the basis to overcome the limitations of current design-alternatives search tools proposing (i) a mathematical formulation for the underlying optimization problem, i.e., determine the Cloud configuration that minimizes the execution costs of the application over a daily time horizon, fulfilling at once QoS and service allocation constraints, and (ii) a software solution able to efficiently solve the resource allocation problem. The tool, codenamed SPACE4Cloud, embodies a hybrid two-level search meta-heuristic for the efficient exploration of the design-alternatives space. The benefits of this approach are demonstrated in the context of an industrial case study. Furthermore, this work also presents results where SPACE4Cloud leads to a cost reduction up to 60%, when compared to a first-principle technique based on utilization thresholds, like the ones typically used in practice. Finally, an extensive scalability analysis indicates that our solution is able to solve large problem instances within a time frame compatible with a fast-paced design process (less than half an hour in the worst case).

show abstract

Section: Related Workmentioning

confidence: 98%

Section: Related Workmentioning

confidence: 99%

Architectural Design of Cloud Applications: A Performance-Aware Cost Minimization Approach

Ciavotta¹,

Gibilisco²,

Ardagna³

et al. 2022

IEEE Trans. Cloud Comput.

Self Cite

View full text Add to dashboard Cite

show abstract

“…In order to enable fully automated search over design space, the SPACE4CLOUD tool was combined with PerOpteryx evolutionary heuristics in a separate study [20]. Evangelinou et al [19,29] further developed such a tool to provide a methodology for migrating existing enterprise applications to Cloud by selecting an optimal deployment topology that takes topology cost and performance into account. To enable faster search, initial solutions for evolutionary algorithm are provided through Mixed-Integer Linear Programming (MILP) algorithm.…”

Section: Resource Locationmentioning

confidence: 99%

QoS-Based Elasticity for Service Chains in Distributed Edge Cloud Environments

Cardellini

Grbac

Nardelli

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

With the emerging IoT and Cloud-based networked systems that rely heavily on virtualization technologies, elasticity becomes a dominant system engineering attribute for providing QoS-aware services to their users. Although the concept of elasticity can introduce significant QoS and cost benefits, its implementation in real systems is full of challenges. Indeed, nowadays systems are mainly distributed, built upon several layers of abstraction, and with centralized control mechanisms. In such a complex environment, controlling elasticity in a centralized manner might strongly penalize scalability. To overcome this issue, we can conveniently split the system in autonomous subsystems that implement elasticity mechanisms and run control policies in a decentralized manner. To efficiently and effectively cooperate with each other, the subsystems need to communicate among themselves to determine elasticity decisions that collectively improve the overall system performance. This new architecture calls for the development of new mechanisms and efficient policies. In this chapter, we focus on elasticity in IoT and Cloud-based systems, which can be geo-distributed also at the edge of the networks, and discuss its engineering perspectives along with various coordination mechanisms. We focus on the design choices that may affect the elasticity properties and provide an overview of some decentralized design patterns related to the coordination of elasticity decisions.

show abstract

“…To address the problem of technology delivery complexity to users, clouds take benefit of brokers; this way, customers can select cost‐effective services with better QoS complying with SLA. () In this paper, we specifically concentrate on MCE security for mission‐critical applications. To do so, we present a system framework and add a module in a cloud broker to log matrix information of security SLA (price, availability, integrity, and confidentiality; cf, Section 3.3.2).…”

Section: Literature Reviewmentioning

confidence: 99%

“…Companies can take the benefit of exploiting several clouds into account instead of one cloud to avoid dependence from one provider and one point of failure. Recently, Amazon and Microsoft advertised the same SLA with 99.95% availability for each; hence, it will be converted into 99.9999% availability in the case of using a combination of both clouds . In MCE, the atomic web services are composited to deliver a user value‐added coarse‐grain service.…”

Section: Introductionmentioning

confidence: 99%

An iterative mathematical decision model for cloud migration: A cost and security risk approach

Shirvani

Sahafi

2017

Softw Pract Exp

View full text Add to dashboard Cite

This paper presents an iterative mathematical decision model for organizations to evaluate whether to invest in establishing information technology (IT) infrastructure on-premises or outsourcing IT services on a multicloud environment. This is because a single cloud cannot cover all types of users' functional/nonfunctional requirements, in addition to several drawbacks such as resource limitation, vendor lock-in, and prone to failure. On the other hand, multicloud brings several merits such as vendor lock-in avoidance, system fault tolerance, cost reduction, and better quality of service. The biggest challenge is in selecting an optimal web service composition in the ever increasing multicloud market in which each provider has its own pricing schemes and delivers variation in the service security level. In this regard, we embed a module in the cloud broker to log service downtime and different attacks to measure the security risk. If security tenets, namely, security service level agreement, such as availability, integrity, and confidentiality for mission-critical applications, are targeted by cybersecurity attacks, it causes disruption in business continuity, leading to financial losses or even business failure. To address this issue, our decision model extends the cost model by using the cost present value concept and the risk model by using the advanced mean failure cost concept, which are derived from the embedded module to quantify cloud competencies. Then, the cloud economic problem is transformed into a bioptimization problem, which minimizes cost and security risks simultaneously. To deal with the combinatorial problem, we extended a genetic algorithm to find a Pareto set of optimal solutions. To reach a concrete result and to illustrate the effectiveness of the decision model, we conducted different scenarios and a small-to-medium business IT development for a 5-year investment as a case study. The result of different implementation shows that multicloud is a promising and reliable solution against IT on-premises deployment.

show abstract

A mixed integer linear programming optimization approach for multi-cloud capacity allocation

Cited by 15 publications

References 40 publications

Architectural Design of Cloud Applications: A Performance-Aware Cost Minimization Approach

Architectural Design of Cloud Applications: A Performance-Aware Cost Minimization Approach

QoS-Based Elasticity for Service Chains in Distributed Edge Cloud Environments

An iterative mathematical decision model for cloud migration: A cost and security risk approach

Contact Info

Product

Resources

About