CloudAware: Empowering context‐aware self‐adaptation for mobile applications

Virtualization technology has revolutionized the mobile network and widely used in fifth‐generation innovation. It is a way of computing that allows dynamic leasing of server capabilities in the form of services like Software as a Service (SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS). The proliferation of these services among the users led to the establishment of large‐scale cloud data centers that consume an enormous amount of electrical energy and results into high metered bill cost and carbon footprint. In this paper, we propose three heuristic models, namely, median migration time, smallest void detection, and maximum fill technique, that can reduce energy consumption with minimal variation in service‐level agreements negotiated. Specifically, we derive the cost of running cloud data center, cost optimization problem, and resource utilization optimization problem. Power consumption model is developed for cloud computing environment focusing on linear relationship between power consumption and resource utilization. A virtual machine (VM) migration technique is considered, focusing on synchronization‐oriented shorter stop‐and‐copy phase. The complete operational steps as algorithms are developed for energy‐aware heuristic models including median migration time, smallest void detection, and maximum fill technique. To evaluate the proposed heuristic models, we conduct experimentations using PlanetLab server data often in ten days and synthetic workload data collected randomly from the similar number of VMs employed in PlanetLab servers. Through evaluation process, we deduce that the proposed approaches can significantly reduce the energy consumption, total VM migration, and host shutdown while maintaining high system performance.

Section: Related Workmentioning

confidence: 99%

Enabling green computing in cloud environments: Network virtualization approach toward 5G support

Verma

Kumar

Kaiwartya

et al. 2018

“…Light‐weight container‐based frameworks have been proposed for service provisioning with minimal service migration time in the MEC environments . Several solutions have been proposed for context‐aware execution of mobile applications at the edge, with focus on bandwidth, latency, and jitter . However, the current solutions fail short in addressing optimality due to the scale and variety of devices and users in the MEC environments.…”

Section: Related Workmentioning

confidence: 99%

Composing network service chains at the edge: A Resilient and adaptive software‐defined approach

Kathiravelu

Roy

Veiga

2018

The scale of data and the demand for high data rates in applications are rapidly increasing over the years. By moving the computing resources closer to the users, edge computing minimizes the latency caused by otherwise pulling the data between the user and a centralized platform. Thus, data centers and computing nodes at the edge continue to overtake or replace traditional centralized cloud platforms in various 5G applications and use cases. Network functions virtualization virtualizes dedicated hardware network middleboxes into virtual network functions (VNFs) and deploys them on top of servers and data centers. Due to the inherent bandwidth cost associated with the distance between the network services and the end user, more and more third‐party VNF providers choose to deploy their services at the edge, rather than in a centralized cloud. A typical network service chaining (NSC) consumes various network services, where one's output becomes the input of another. As network services are often latency sensitive, NSC needs to be carried out, maintaining proximity to the user as well as among the services composing the NSC. In this paper, we present the design of a resilient and adaptive framework, following an approach inspired by software‐defined networking, for the remote users to construct network service chains in the mobile and edge computing environments. Our evaluations highlight the efficiency and optimality of the framework in finding the best fit among several third‐party edge VNFs in deploying the user network service chains, abiding by various user policies.

“…The overall quality of the service is determined by several parameters, ie, communication latency and bandwidth, CPU time available for mobile services, acceptance rate of new connections, seamless handovers/migrations, and resilience to failures. Such parameters are more or less directly affected by transmission and computing mechanisms (eg, priority classes and queue management for network packets and scheduling and preemption for software jobs), placement decisions (eg, number of communication hops), and operational decisions (eg, rerouting of packet streams, live migrations of applications and services, and resource allocation strategies); we briefly indicate these aspects under the generic term of “QoS management.” Effective QoS management is essential to guarantee the required performance levels during the whole service lifetime, so as to effectively support smart manufacturing, e‐health, energy, automotive, and media and entertainment industries.…”

Section: Quality Of Service Versus Energy Efficiency In Edge Computingmentioning

confidence: 99%

Balancing QoS and power consumption in programmable 5G infrastructures

Carrega

Portomauro

Repetto

et al. 2018

The advent of fifth generation is promising to push far more intelligence than today to the network boundary, hence boosting novel computing models based on fog/edge paradigms. The need for proximity in computation, coupled with various forms of mobility, will be responsible for dynamic shifting of workload within the system, with large fluctuations in resource usage. This eventually turns into poor energy efficiency of the whole infrastructure. However, improving efficiency usually deteriorates quality of service, hence the dilemma about how to balance these two contrasting aspects. In this paper, we propose a framework that leverages the increasing programmability of ICT infrastructures to pursue a linear relationship between power consumption and workload, while safeguarding quality of service. Our approach is based on workload consolidation and extensions to existing cloud management software. We collected both real measurements from an experimental testbed and performance analysis from simulations to evaluate the consolidation strategy in more complex environments.