arXiv subjects

Jaafar M. H. Elmirghani

Publications and source records attributed to Jaafar M. H. Elmirghani.

At least 73 records · Page 4Linked to original sources

arXiv2021

Energy Minimized Federated Fog Computing over Passive Optical Networks

The rapid growth of time-sensitive applications and services has driven enhancements to computing infrastructures. The main challenge that needs addressing for these applications is the optimal placement of the end-users demands to reduce the total power consumption and delay. One of the widely adopted paradigms to address such a challenge is fog computing. Placing fog units close to end-users at the edge of the network can help mitigate some of the latency and energy efficiency issues. Compared to the traditional hyperscale cloud data centres, fog computing units are constrained by computational power, hence, the capacity of fog units plays a critical role in meeting the stringent demands of the end-users due to intensive processing workloads. In this paper, we aim to optimize the placement of virtual machines (VMs) demands originating from end-users in a fog computing setting by formulating a Mixed Integer Linear Programming (MILP) model to minimize the total power consumption through the use of a federated architecture made up of multiple distributed fog cells. The obtained results show an increase in processing capacity in the fog layer and a reduction in the power consumption by up to 26% compared to the Non-Federated fogs network.

cs.NI↗

arXiv2021

Energy-Efficient Distributed Machine Learning in Cloud Fog Networks

Massive amounts of data are expected to be generated by the billions of objects that form the Internet of Things (IoT). A variety of automated services such as monitoring will largely depend on the use of different Machine Learning (ML) algorithms. Traditionally, ML models are processed by centralized cloud data centers, where IoT readings are offloaded to the cloud via multiple networking hops in the access, metro, and core layers. This approach will inevitably lead to excessive networking power consumptions as well as Quality-of-Service (QoS) degradation such as increased latency. Instead, in this paper, we propose a distributed ML approach where the processing can take place in intermediary devices such as IoT nodes and fog servers in addition to the cloud. We abstract the ML models into Virtual Service Requests (VSRs) to represent multiple interconnected layers of a Deep Neural Network (DNN). Using Mixed Integer Linear Programming (MILP), we design an optimization model that allocates the layers of a DNN in a Cloud/Fog Network (CFN) in an energy efficient way. We evaluate the impact of DNN input distribution on the performance of the CFN and compare the energy efficiency of this approach to the baseline where all layers of DNNs are processed in the centralized Cloud Data Center (CDC).

cs.NI↗

arXiv2021

Energy-Efficient AI over a Virtualized Cloud Fog Network

Deep Neural Networks (DNNs) have served as a catalyst in introducing a plethora of next-generation services in the era of Internet of Things (IoT), thanks to the availability of massive amounts of data collected by the objects on the edge. Currently, DNN models are used to deliver many Artificial Intelligence (AI) services that include image and natural language processing, speech recognition, and robotics. Accordingly, such services utilize various DNN models that make it computationally intensive for deployment on the edge devices alone. Thus, most AI models are offloaded to distant cloud data centers (CDCs), which tend to consolidate large amounts of computing and storage resources into one or more CDCs. Deploying services in the CDC will inevitably lead to excessive latencies and overall increase in power consumption. Instead, fog computing allows for cloud services to be extended to the edge of the network, which allows for data processing to be performed closer to the end-user device. However, different from cloud data centers, fog nodes have limited computational power and are highly distributed in the network. In this paper, using Mixed Integer Linear Programming (MILP), we formulate the placement of DNN inference models, which is abstracted as a network embedding problem in a Cloud Fog Network (CFN) architecture, where power savings are introduced through trade-offs between processing and networking. We study the performance of the CFN architecture by comparing the energy savings when compared to the baseline approach which is the CDC.

cs.NI↗

arXiv2021

Safety Analysis for Laser-based Optical Wireless Communications: A Tutorial

Light amplification by stimulated emission of radiation (laser) sources have many advantages for use in high data rate optical wireless communications. In particular, the low cost and high-bandwidth properties of laser sources such as vertical-cavity surface-emitting lasers (VCSELs) make them attractive for future indoor optical wireless communications. In order to be integrated into future indoor networks, such lasers should conform to eye safety regulations determined by the international electrotechnical commission (IEC) standards for laser safety. In this paper, we provide a detailed study of beam propagation to evaluate the received power of various laser sources, based on which as well as the maximum permissible exposure (MPE) defined by the IEC 60825-1:2014 standard, we establish a comprehensive framework for eye safety analyses. This framework allows us to calculate the maximum allowable transmit power, which is crucial in the design of a reliable and safe laser-based wireless communication system. Initially, we consider a single-mode Gaussian beam and calculate the maximum permissible transmit power. Subsequently, we generalize this approach for higher-mode beams. It is shown that the M-squared-based approach for analysis of multimode lasers ensures the IEC eye safety limits, however, in some scenarios, it can be too conservative compared to the precise beam decomposition method. Laser safety analyses with consideration of optical elements such as lens and diffuser, as well as for VCSEL array have been also presented. Skin safety, as another significant factor of laser safety, has also been investigated in this paper. We have studied the impacts of various parameters such as wavelength, exposure duration and the divergence angle of laser sources on the safety analysis by presenting insightful results.

cs.IT↗

arXiv2021

Power Minimization in Vehicular Cloud Architecture

Modern vehicles equipped with on-board units (OBU) are playing an essential role in the smart city revolution. The vehicular processing resources, however, are not used to their fullest potential. The concept of vehicular clouds is proposed to exploit the underutilized vehicular resources to supplement cloud computing services to relieve the burden on cloud data centers and improve quality of service. In this paper we introduce a vehicular cloud architecture supported by fixed edge computing nodes and the central cloud. A mixed integer linear programming (MLP) model is developed to optimize the allocation of the computing demands in the distributed architecture while minimizing power consumption. The results show power savings as high as 84% over processing in the conventional cloud. A heuristic with performance approaching that of the MILP model is developed to allocate computing demands in real time.

cs.NI↗

arXiv2021

Resilient Architectures for Free Space Optical Wireless Interconnection Systems

In this paper, we propose the use of two Passive Optical Network (PON)-based network architectures to connect free-space Optical Wireless Communication (OWC) Access Points (APs) within a room with multiple users. We optimize through a Mixed Linear Integer Programming (MILP) model the assignment of mobile OWC users to more than one AP to improve the resilience of the fronthaul network, i.e the OWC system and the wired network linked to APs, and study the impact of users distribution and channel characteristics.

cs.NI↗

arXiv2021

Optimized Energy Efficient Virtualization and Content Caching in 5G Networks

Network function virtualization (NFV) and content caching are two promising technologies that hold great potential for network operators and designers. This paper optimizes the deployment of NFV and content caching in 5G networks and focuses on the associated power consumption savings. In addition, it introduces an approach to combine content caching with NFV in one integrated architecture for energy aware 5G networks. A mixed integer linear programming (MILP) model has been developed to minimize the total power consumption by jointly optimizing the cache size, virtual machine (VM) workload, and the locations of both cache nodes and VMs. The results were investigated under the impact of core network virtual machines (CNVMs) inter-traffic. The result show that the optical line terminal (OLT) access network nodes are the optimum location for content caching and for hosting VMs during busy times of the day whilst IP over WDM core network nodes are the optimum locations for caching and VM placement during off-peak time. Furthermore, the results reveal that a virtualization-only approach is better than a caching-only approach for video streaming services where the virtualization-only approach compared to caching-only approach, achieves a maximum power saving of 7% (average 5%) when no CNVMs inter-traffic is considered and 6% (average 4%) with CNVMs inter-traffic at 10% of the total backhaul traffic. On the other hand, the integrated approach has a maximum power saving of 15% (average 9%) with and without CNVMs inter-traffic compared to the virtualization-only approach, and it achieves a maximum power saving of 21% (average 13%) without CNVMs inter-traffic and 20% (average 12%) when CNVMs inter-traffic is considered compared with the caching-only approach. In order to validate the MILP models and achieve real-time operation in our approaches, a heuristic was developed.

cs.NI↗

arXiv2021

Disaggregation for Energy Efficient Fog in Future 6G Networks

We study the benefits of adopting server disaggregation in the fog computing tier by evaluating energy efficient placement of interactive apps in a future fog 6G network. Using a mixed integer linear programming (MILP) model, we compare the adoption of traditional server (TS) and disaggregated server (DS) architectures in a fog network comprising of selected fog computing sites in the metro and access networks. Relative to the use of TSs, our results show that the adoption of DS improves the energy efficiency of the fog network and enables up to 18% reduction in total fog computing power consumption. More instances of interactive fog apps are provisioned in a fog network that is implemented over a network topology with high delay penalty. This ensures that minimal delay is experienced by distributed users. Our result also shows that the proximity of fog computing sites such as metro-central offices and radio cell sites to geo-distributed users of interactive fog applications make them important edge locations for provisioning moderately delay sensitive fog apps. However, fog applications with more stringent delay thresholds require in situ processing at directly connected radio cell sites or at the location of the requesting users. Finally, we propose a heuristic for energy efficient and delay aware placement of interactive fog apps in a fog network which replicates the trends observed during comprehensive analysis of the exact results obtained by solving the MILP model formulated in this paper. Our results and proposed MILP and heuristic provide a good reference and tool for fog network design and deployment.

cs.NI↗

arXiv2021

Energy Efficient Placement of Workloads in Composable Data Center Networks

This paper studies the energy efficiency of composable datacentre (DC) infrastructures over network topologies. Using a mixed integer linear programming (MILP) model, we compare the performance of disaggregation at rack-scale and pod-scale over selected electrical, optical and hybrid network topologies relative to a traditional DC. Relative to a pod-scale DC, the results show that physical disaggregation at rack-scale is sufficient for optimal efficiency when the optical network topology is adopted and resource components are allocated in a suitable manner. The optical network topology also enables optimal energy efficiency in composable DCs. The paper also studies logical disaggregation of traditional DC servers over an optical network topology. Relative to physical disaggregation at rack-scale, logical disaggregation of server resources within each rack enables marginal fall in the total DC power consumption (TDPC) due to improved resource demands placement. Hence, an adaptable composable infrastructure that can support both in memory (access) latency sensitive and insensitive workloads is enabled. We also conduct a study of the adoption of micro-service architecture in both traditional and composable DCs. Our results show that increasing the modularity of workloads improves the energy efficiency in traditional DCs, but disproportionate utilization of DC resources persists. A combination of disaggregation and micro-services achieved up to 23% reduction in the TDPC of the traditional DC by enabling optimal resources utilization and energy efficiencies. Finally, we propose a heuristic for energy efficient placement of workloads in composable DCs which replicates the trends produced by the MILP model formulated in this paper.

cs.NI↗

arXiv2020

Cloud Fog Architectures in 6G Networks

Prior to the advent of the cloud, storage and processing services were accommodated by specialized hardware, however, this approach introduced a number of challenges in terms of scalability, energy efficiency, and cost. Then came the concept of cloud computing, where to some extent, the issue of massive storage and computation was dealt with by centralized data centers that are accessed via the core network. The cloud has remained with us thus far, however, this has introduced further challenges among which, latency and energy efficiency are of the pinnacle. With the increase in embedded devices intelligence came the concept of the Fog. The availability of massive numbers of storage and computational devices at the edge of the network, where some are owned and deployed by the end-users themselves but most by service operators. This means that cloud services are pushed further out from the core towards the edge of the network, hence reduced latency is achieved. Fog nodes are massively distributed in the network, some benefit from wired connections, and others are connected via wireless links. The question of where to allocate services remains an important task and requires extensive attention. This chapter introduces and evaluates cloud fog architectures in 6G networks paying special attention to latency, energy efficiency, scalability, and the trade-offs between distributed and centralized processing resources.

cs.NI↗

arXiv2020

Resource Allocation in 6G Optical Wireless Systems

The abundant optical spectrum is a promising part of the electromagnetic spectrum for 6G communication systems. The visible light spectrum which is a part of the optical spectrum, can be used to provide communication and illumination simultaneously. Visible light communication (VLC) systems have been widely researched, however, little work has focused on the area of multiple access. This chapter studies wavelength division multiple access (WDMA) techniques in VLC systems to support multiple users. In addition, the optimization of resource allocation is considered in this chapter by developing a mixed-integer linear programming (MILP) model that can be used to maximize the signal to noise and interference ratio (SINR) while supporting multiple users. The optimized resource allocation results in the best assignment of access points (APs) and wavelengths to users. Different indoor environments such as office, data center and aircraft cabins are evaluated in this chapter. A laser diode (LD) with four wavelengths (red, green, yellow and blue) is used to provide high bandwidth for communication and white light for illumination. Also, an angle diversity receiver (ADR) is utilized to receive signals and reduce noise and interference by exploiting the spatial domain.

eess.SP↗

arXiv2020

Energy Efficient Fog based Healthcare Monitoring Infrastructure

Recent advances in mobile technologies and cloud computing services have inspired the development of cloud-based real-time health monitoring systems. However, the transfer of health-related data to the cloud contributes to the burden on the networking infrastructures, leading to high latency and increased power consumption. Fog computing is introduced to relieve this burden by bringing services to the users proximity. This study proposes a new fog computing architecture for health monitoring applications based on a Gigabit Passive Optical Network (GPON) access network. An Energy-Efficient Fog Computing (EEFC) model is developed using Mixed Integer Linear Programming (MILP) to optimize the number and location of fog devices at the network edge to process and analyze the health data for energy-efficient fog computing. The performance of the EEFC model at low data rates and high data rates health applications is studied. The outcome of the study reveals that a total energy saving of 36% and 52% are attained via processing and analysis the health data at the fog in comparison to conventional processing and analysis at the central cloud for low data rate application and high data rate application, respectively. We also developed a real-time heuristic; Energy Optimized Fog Computing (EOFC) heuristic, with energy consumption performance approaching the EEFC model. Furthermore, we examined the energy efficiency improvements under different scenarios of devices idle power consumption and traffic volume.

cs.NI↗

arXiv2020

Blackout Resilient Optical Core Network

A disaster may not necessarily demolish the telecommunications infrastructure, but instead it might affect the national grid and cause blackouts, consequently disrupting the network operation unless there is an alternative power source(s). In this paper, power outages are considered, and the telecommunication network performance is evaluated during a blackout. Two approaches are presented to minimize the impact of power outage and maximize the survival time of the blackout node. A mixed integer linear programming (MILP) model is developed to evaluate the network performance under a single node blackout scenario. The model is used to evaluate the network under the two proposed scenarios. The results show that the proposed approach succeeds in extending the network life time while minimizing the required amount of backup energy.

cs.NI↗

arXiv2020

Optimization of Processing Allocation in Vehicular Edge Cloud based Architecture

Vehicular edge computing is a new distributed processing architecture that exploits the revolution in the processing capabilities of vehicles to provide energy efficient services and low delay for Internet of Things (IoT)-based systems. Edge computing relies on a set of distributed processing nodes (i.e. vehicles) that are located close to the end user. In this paper, we consider a vehicular edge cloud (VEC) consisting of a set of vehicle clusters that form a temporal vehicular cloud by combining their computational resources in the cluster. We tackle the problem of processing allocation in the proposed vehicular edge architecture by developing a Mixed Integer Linear Programming (MILP) optimization model that jointly minimizes power consumption, propagation delay, and queuing delay. The results show that the closer the processing node (PN) is to the access point (AP), the lower the power consumption and delay, as the distance and number of hops affect the propagation delay and queuing delay. However, the queuing delay at the AP becomes a limiting factor when it operates at a low service rate compared to the traffic arrival rate. Thus, processing tasks at the vehicular nodes (VN) was avoided whenever the objective function included queueing delay and the AP operated at a low service rate. Increase in the AP service rate results in a lower queuing delay and better VN utilization.

cs.NI↗

arXiv2020

Energy Efficient Processing Allocation in Opportunistic Cloud-Fog-Vehicular Edge Cloud Architectures

This paper investigates distributed processing in Vehicular Edge Cloud (VECs), where a group of vehicles in a car park, at a charging station or at a road traffic intersection, cluster and form a temporary vehicular cloud by combining their computational resources in the cluster. We investigated the problem of energy efficient processing task allocation in VEC by developing a Mixed Integer Linear Programming (MILP) model to minimize power consumption by optimizing the allocation of different processing tasks to the available network resources, cloud resources, fog resources and vehicular processing nodes resources. Three dimensions of processing allocation were investigated. The first dimension compared centralized processing (in the central cloud) to distributed processing (in the multi-layer fog nodes). The second dimension introduced opportunistic processing in the vehicular nodes with low and high vehicular node density. The third dimension considered non-splittable tasks (single allocation) versus splittable tasks (distributed allocation), representing real-time versus non real-time applications respectively. The results revealed that a power savings up to 70% can be achieved by allocating processing to the vehicles. However, many factors have an impact on the power saving such the vehicle processing capacities, vehicles density, workload size, and the number of generated tasks. It was observed that the power saving is improved by exploiting the flexibility offered by task splitting among the available vehicles.

cs.NI↗

arXiv2020

VM placement over WDM-TDM AWGR PON Based Data Centre Architecture

Passive optical networks (PON) can play a vital role in data centres and access fog solutions by providing scalable, cost and energy efficient architectures. This paper proposes a Mixed Integer Linear Programming (MILP) model to optimize the placement of virtual machines (VMs) over an energy efficient WDM-TDM AWGR PON based data centre architecture. In this optimization, the use of VMs and their requirements affect the optimum number of servers utilized in the data centre when minimizing the power consumption and enabling more efficient utilization of servers is considered. Two power consumption minimization objectives were examined for up to 20 VMs with different computing and networking requirements. The results indicate that considering the minimization of the processing and networking power consumption in the allocation of VMs in the WDM-TDM AWGR PON can reduce the networking power consumption by up to 70% compared to the minimization of the processing power consumption.

cs.NI↗

arXiv2020

Caching Video-on-Demand in Metro and Access Fog Data Centres

This paper examines the utilization of metro fog data centres and access fog datacentres with integrated solar cells and Energy Storage Devices (ESDs) to assist cloud data centres in caching Video-on-Demand content and hence, reduce the networking power consumption. A Mixed Integer Linear Programming (MILP) model is used to optimize the delivery of the content from cloud, metro fog, or access fog datacentres. The results for a range of data centre parameters show that savings by up to 38% in the transport network power consumption can be achieved when VoD is optimally served from fully renewable-powered cloud or metro fog data centres or from access fog data centres with 250 m2 solar cells. Additional 8% savings can be achieved when using ESDs of 100 kWh capacity in the access fog data centres.

cs.NI↗

arXiv2020

NOMA Visible Light Communication System with Angle Diversity Receivers

In this paper, a non-orthogonal multiple access (NOMA) visible light communication (VLC) system is investigated. The system uses angle diversity receivers (ADRs) to provide high data rates. The ADR has 4 branches, each directed to a different direction. An 8m x 4m sized room is modelled to study the resource allocation to users according to their channel conditions to maximize the data rate. The results show that using ADRs improves the data rate by an average of 35% compared to a system using wide FOV receivers.

eess.SP↗