Daniel M. Batista

Mapping virtual networks onto substrate networks

Network virtualization is a promising technique for building the Internet of the future since it enables the low cost introduction of new features into network elements. An open issue in such virtualization is how to effect an efficient mapping of virtual network elements onto those of the existing physical network, also called the substrate network. Mapping is an NP-hard problem and existing solutions ignore various real network characteristics in order to solve the problem in a reasonable time frame. This paper introduces new algorithms to solve this problem based on 0–1 integer linear programming, algorithms based on a whole new set of network parameters not taken into account by previous proposals. Approximative algorithms proposed here allow the mapping of virtual networks on large network substrates. Simulation experiments give evidence of the efficiency of the proposed algorithms.

Optimal Mapping of Virtual Networks

Network virtualization is a promising technique for building the Internet of the future since it enables the low cost introduction of new features into network elements. An open issue in virtualization is how to search for an efficient mapping of virtual network elements onto those of the existing physical network, also called the substrate network. Mapping is an NP-hard problem and existing solutions ignore various real network characteristics in order to solve problem in a reasonable time frame. This paper introduces two new algorithms for the solution of the mapping problem, both based on 0--1 integer programming, for the solution of the mapping problem which consider a whole new set of network parameters not taken into account by previous proposals. Simulation experiments confirm the efficiency of the proposed algorithms.

Green virtualized networks

The Future Internet demands energy efficient communication to cope with the ever increasing power consumption. Virtualization techniques have proved to be effective in reducing power consumption of network devices. An open issue in virtualization for green networking is the search for an energy-efficient mapping of virtual networks onto physical networks. This paper introduces a new model for the mapping of virtual networks which aims at reducing the energy consumption. This model is based on an integer linear programming formulation and several parameters, corresponding to characteristic of real networks, are considered. Simulation results attest the efficacy of the proposal.

Energy-Efficient Virtual Machines Placement

Energy efficiency on computer systems is a topic that is gaining a lot of interest. Even more in the cloud computing era, where data centres consumption corresponds to near 1.5% of total world wide power consumption. In this paper we present two novel approaches for virtual machines (VMs) placement consolidation. The two approaches aim to maximize the placed VMs on a host and therefore minimize the number of hosts used on a cloud computing environment. The first proposed approach is based on the Knapsack problem and the second one is based on an Evolutionary Computation heuristic. Both strategies have shown consumed energy reduction starting from 40.33% and up to 92.21% compared to a strategy that does not consider energy efficiency.

A green network-aware VMs placement mechanism

Data centers power consumption corresponds to near 2% of the total world wide power consumption, with constantly increasing greenhouse effect and CO2 footprints. Virtualization techniques improve the efficiency of data centers infrastructure sharing a same physical hardware among several Virtual Machines (VMs). An efficient VM placement can minimize even further the hardware and energy needs. In contrast to existing VM placement algorithms that usually focus on a single resource or assumes that resources demands are deterministic, this paper proposes and compares four energy-aware algorithms that consider multiple stochastic resources, including network bandwidth. We first formulate the problem as a multi objective optimization problem with stochastic resources and we present two algorithms based on this approach. We also formulate the problem as an evolutionary computation problem and we present two algorithms based on this approach. The objective is a joint strategy: minimize the required hardware to maximize the allocated VMs satisfying the resource requirements. Through simulations, we compare our algorithms using real VMs workloads from the PlanetLab project and showed the significant improvements on power consumption and network utilization. In average, the algorithms reduce power consumption by 87.90% and the network utilization by 9.94%.

A dynamic frame slotted ALOHA anti-collision algorithm for the internet of things

This paper proposes a novel anti-collision algorithm for passive RFID systems to be applied in the Internet of Things environments. The proposed algorithm is based on a Dynamic Frame Slotted ALOHA algorithm and is called NEDFSA. In the algorithm, tags are randomly assigned to time slots of a frame and when two or more tags collide in a slot, the collisions will be resolved by another DFSA algorithm while the other tags in the next slots will wait. The proposed algorithm uses an estimation mechanism to adjust the initial frame length to a value close to the number of tags, based on a power of two. The advantages of the proposal are that it retains the simplicity of DFSA, and its performance is not affected by the number of tags. Simulation results in ns-2 show that the efficiency of NEDFSA is up to 27% better than Q Algorithm. Also, the results show that the graphs plotting the efficiency of the algorithm are nearly horizontal when the number of tags increases from 100 to 1800, which confirms the high scalability of NEDFSA.

A RFID QoS mechanism for IoT tracking applications

In this paper, we propose a mechanism that reduces useless responses for RFID systems in the Internet of Things (IoT) environment, for tracking applications. The mechanism achieves its objectives by reducing the number of packets exchanged between readers and tags. We analyze the behavior of our proposal by considering the average number of identification rounds. With extensive simulations using an RFID module for the ns-2 simulator, we show the benefit of the proposed mechanism. When compared to the Pure Q Algorithm and Binary Tree Slotted Aloha, our mechanism reduces the number of packets up to 43%, which is a good result in terms of performance of IoT applications and energy consumption of the devices used in the communications.

An Ns-2 Module for Simulating Passive RFID Systems

This paper presents the design and validation of an RFID module for the Network Simulator 2 which implements the ISO/IEC 18000-6C Class 1 Generation 2 standard, Schoute and Eom-Lee anti-collisions protocols in order to evaluate the performance of RFID systems and Internet of Things environments. The module includes mechanisms for the definition of parameters related to the standards and for QoS provision. Moreover, the implementation is open source and intended to be a reference tool for evaluating RFID protocols for anti-collision algorithms, security, privacy and other desired resources. Brief results obtained with the module showing the benefits of a QoS mechanism proposed by the authors are also presented. The module is important to the network community and this paper provides a scientific methodology attesting its functionality and conformity with existing standards and protocols.

Approximated algorithms for mapping virtual networks on network substrates

Network virtualization is a promising technique for building the Internet of the future since it enables the introduction of new features into network elements at low cost. An open issue in virtualization is how to search for an efficient mapping of virtual network elements onto those of the existing physical network. Mapping is an NP-hard problem and existing solutions take long time to find a solution. This paper presents four new approximated algorithms based on two integer linear programming formulations that runs fast and, also, consider various real network characteristics, which is neglected by other proposals in the literature.

Consolidation of VMs to Improve Energy Efficiency in Cloud Computing Environments

Improvement of energy efficiency in IT is an important research topic nowadays. The reduction of operational costs, generated heat and environment impact are some of the reasons for this. Thanks to the advent of cloud computing, it is possible to improve energy efficiency in data centers by running various virtual machines in a single physical machine. However, the cloud providers generally invest in performance, not energy efficiency. This paper focuses on the problem of an energy efficient initial VM placement, and describes three new algorithms for this problem, one based on the First Fit Decreasing algorithm, and the other two based on the Best Fit Decreasing algorithm. They are compared with other algorithms in the literature, and a reduction of power consumption up to 3.24% was observed, as well a reduction of execution time in several orders of magnitude. Scripts used to analyze traces publicly provided by Google are another contribution of the paper, since they are useful for those working in mechanisms for cloud computing.