Multi Objective Virtual Machine Allocation in Cloud Data Centers

Portaluri, Giuseppe; Giordano, Stefano

doi:10.1109/cloudnet.2016.8

Cited by 10 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also some heuristic algorithms for VNF placement have been already proposed, as in [14]. In [15], the authors proposed a virtual machine allocator for cloud computing data center. They allocate a set of virtual machines on servers that are interconnected through a three-tier fattree network topology.…”

Section: Related Workmentioning

confidence: 99%

“…(12) puts a limit to the number of fibers for each link. (13) ensures that there is at least one physical path for each optical tunnel of the IP/MPLS layer, while (14) and (15) map the the path on the physical layer for each IP/MPLS tunnels with different source and destination nodes. (16) creates a tunnel between two nodes only if there exist traffic between the NFs associated to the MPLS nodes.…”

Section: A Complete Milp Formulationmentioning

confidence: 99%

See 1 more Smart Citation

Dynamic programming of network slices in software-defined metro-core optical networks

Troia

Cibari

Alvizu

et al. 2020

Optical Switching and Networking

View full text Add to dashboard Cite

Nowadays networks are the basis of our communication providing a great number of services. As a consequence, the traffic is increasing and there is a growing demand for new services that require stringent constraints on capacity, latency and jitter to provide an appropriate Quality of Service (QoS) to end users. In order to cope with these requirements, network infrastructure needs to evolve from a static and closed architecture towards a more scalable, dynamic and agile one. Software-Defined Networking and Network Function Virtualization allow to provide different services, each one with its own QoS constraints, independent and secure, thanks to the network slicing concept, the main subject of this work. Network slicing allows to segment the underlying physical network into different logical networks to provide data transport customized to specific services. In this paper, we propose two mathematical models able to dynamically provision network slices on the physical network, complying with their QoS requirements for their instantiation and routing of traffic. The proposed models aim at minimizing a linear combination of probability of blocking traffic requests, energy consumption of physical network devices and interruption of service due to the reconfiguration of the slices. Taking advantage of traffic signatures from a city's mobile network, the goal is to predict how and when to reconfigure slices already deployed in the network with the aim to optimize the resource allocation in the underlying physical network.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: A Complete Milp Formulationmentioning

confidence: 99%

Dynamic programming of network slices in software-defined metro-core optical networks

Troia

Cibari

Alvizu

et al. 2020

Optical Switching and Networking

View full text Add to dashboard Cite

show abstract

“…The assignment problem is a combination with a non-convex shape, it is a versatile problem is an NP-difficult packing variant. For these reasons, we use special heuristics on addresses, so-called multi-objective genetic algorithms (MOGAs), to inspire the natural process of problem evolution, MOGAs are often effectively complex of the problem [9].…”

Section: Related Workmentioning

confidence: 99%

Elastic Virtualized Network Security Based on Multi-Tenant using Progressive Route Scheduling in a Cloud Data Center

U.,

2020

IJEAT

View full text Add to dashboard Cite

Information security in the cloud has become a serious problem on the Internet. There are safety standards created by the resource in traditional computing. The public cloud providers, secure transfer, and the use of public resources due to the availability of the Internet is the cloud facing distributed data centers. Providing the required level of security, approaching adopting various security components such as security watchdog, IDS / IPS system, security framework, access control framework, security management and so on. In this proposed system to improve the data transmission security to introduce proposed algorithm Elastic virtualized network security (eVNS) for extended security service which utilizes the progressive route scheduling (PRS) and promotes the security consumption for Inbound rule based on the virtualized security load balancer. We have presented a virtual cloud for a network that is secure and robust security group protocols by any compromised or faulty node in the network. In contrast to the traditional snapshot aggregation approach in data centers, the eVNS-PRS proposed algorithm resembles the unicasting it's sensed information to create a new target security group based on the routing table. This makes the system more fault-tolerant and increases the availability of information in the network. Simulations performed with the proposed algorithm have demonstrated its effectiveness.

show abstract

“…In the work [11] Portaluri, Giuseppe et al applied MOGA algorithm to reduce the power consumption in data center. The authors evaluate the proposed algorithm by combining between different numbers of auxiliary objectives which affected negatively on quality of results.…”

Section: Related Workmentioning

confidence: 99%

A new task scheduling approach based on Spacing Multi-Objective Genetic algorithm in cloud

Bey¹,

Belgacem²,

Nacer³

2018

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

The dazzling progress in information and communication technologies, contributed significantly to the emergence of cloud computing paradigm, where it promotes prosperity in all fields of human activity, especially in business. Furthermore, manage the resources and use in ways that sharing with large number of users, consider as one of the challenges facing cloud computing environment today. Because cloud processes a huge tasks, which require the employment of scheduling techniques to handle and monitor the resources in an optimal, flexible and dynamic manner. In this paper, we review a new approach called Spacing-MOGA based on spacing distance to rank no-dominate solutions. It aims mainly to minimize both the makespan and cost of execution tasks on virtual machines (VMs). As well, we study its impact on the availability of resources. Experimental results show that S-MOGA is better than Max-min, PSO and MOGA methods, especially as it minimizes the number of active VMs.

show abstract

Multi Objective Virtual Machine Allocation in Cloud Data Centers

Cited by 10 publications

References 12 publications

Dynamic programming of network slices in software-defined metro-core optical networks

Dynamic programming of network slices in software-defined metro-core optical networks

Elastic Virtualized Network Security Based on Multi-Tenant using Progressive Route Scheduling in a Cloud Data Center

A new task scheduling approach based on Spacing Multi-Objective Genetic algorithm in cloud

Contact Info

Product

Resources

About