On the Modeling of Reliability in Extreme Edge Computing Systems

Allahham, Mhd Saria; Mohamed, Amr; Erbad, Aiman; Hassanein, Hossam S.

doi:10.1109/iccspa55860.2022.10019108

Cited by 6 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Extreme Edge Computing (EEC) is the bottom layer of the edge computing spectrum and aims to offload tasks to edge devices owned by consumers and users rather than edge servers [6], [5]. Some authors have named this layer as the Mist or the Things [5], which extends processing to the network edge, including the sensor and actuator devices.…”

Section: A Extreme Edge Computing Environmentmentioning

confidence: 99%

“…This increased use of technology has led to a higher demand for computational resources that are highly available, which, in turn, results in an increase in the number of data centers at the edge of the network, significantly amplifying CO 2 emissions and server maintenance costs. To propose more sustainable solutions in computing, it is necessary to shift towards Extreme Edge Computing (EEC) [4], [5], tak-ing advantage of the massive computing resources available at Extreme Edge Devices (EED) [6] such as laptops, PCs, tablets, smartphones, and connected vehicles, rather than increasing the number of high-capacity servers at the edge of the network. EEC offers computing services much closer to end-users, which can reduce delays and improve the quality of service.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Objective Deep Reinforcement Learning for Efficient Workload Orchestration in Extreme Edge Computing

Safavifar,

Gyamfi,

Mangina

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Workload orchestration at the edge of the network has become increasingly challenging with the ever-increasing penetration of resource demanding mobile, and heterogeneous devices offering low latency services. Literature has addressed this challenge assuming the availability of multi-access Mobile Edge Computing (MEC) servers and placing the computing tasks related to such services on the MEC servers. However, to develop a more sustainable and energy-efficient computing paradigm, for applications operating in stochastic environments with unpredictable workloads, it is essential to minimize the MEC servers' usage, and utilize the available resource-constrained edge devices, to keep the resourceful servers idle for handling any unpredictable larger workload. In this paper, we proposed DEWOrch, a deep reinforcement Learning algorithm for efficient workload orchestration. DEWOrch's aim is to increase the utilization of resource-constrained edge devices and minimize resource waste for more sustainable and energy efficient computing solution. This model is evaluated in an Extreme Edge Computing environment, where no MEC servers is available and only edge devices with constrained capacity are used to perform tasks. The results show that DEWOrch outperforms the state-of-the-art methods by around 50% decrease in resource waste while improved task success rate, and decreased energy consumption per task in most scenarios.

show abstract

Section: A Extreme Edge Computing Environmentmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Multi-Objective Deep Reinforcement Learning for Efficient Workload Orchestration in Extreme Edge Computing

Safavifar,

Gyamfi,

Mangina

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…) As such, computing performance metrics based on Eq. 1 can be done by utilizing the Law of the Unconscious Statistician (LOTUS) where (13) where both integrals are transcendental and cannot be simplified further. This is mainly due to the similarity of these expressions to the Gaussian error function, erf(•).…”

Section: Performance Analysis Under Uniform Job Incentivesmentioning

confidence: 99%

“…As a consequence, a new layer on the edge-cloud continuum is conceivable, positioned beyond the last mile of current EC, at the extreme edge. This layer is known as Extreme Edge Computing (XEC) [1], [13].…”

Section: Introductionmentioning

confidence: 99%

Incentive-Vacation Queueing in Extreme Edge Computing: An Analytical Reward-Based Framework

Azmy,

Zorba,

Hassanein

2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

Edge Computing (EC) emerged to address the cloud's shortcomings in meeting demand and latency requirements, leading to a shift in computation closer to the end-user. Extreme Edge Computing (XEC) extends this approach by utilizing nearby user-owned computational resources to support latencysensitive applications in a distributed manner. In this study, we introduce Reward Edge Computing (REC), a variant of XEC, where service providers recruit user devices for infrastructure support, offering rewards in return. We explore the use of Incentive-Vacation Queueing (IVQ) to manage REC and analyze both its long-term and short-term performance. Our analysis focuses on the choice of an Incentive-Vacation Function (IVF), a contractual function between workers and service providers, proposing a tunable model favoring either party. We provide closed-form expressions for long-term worker behavior under uniform workload pricing and analyze the system's overall short-term operation, including the time a worker spends in the system. REC and IVQ aim to commodify computational resources for edge services, akin to sharing economy models like Uber and Airbnb, utilizing user-owned infrastructure.

show abstract

“…These graphs showcase the entirety of processing and communication resources available across different levels within the system. In Figure 2(b), the graph's vertices represent the different cloud's Extreme Edge Devices [40] (hereafter, referred to as nodes), while the edges indicate the connection between two nodes. Because network nodes u ∈ U differ in their computing resources (e.g., CPU and Memory), the max amount of resources of type k in node u is defined as a u (k).…”

Section: ) Computing and Network Resourcesmentioning

confidence: 99%

ECP: Error-Aware, Cost-Effective and Proactive Network Slicing Framework

Aboeleneen,

Abdellatif,

Erbad

et al. 2024

IEEE Open J. Commun. Soc.

View full text Add to dashboard Cite

Recent advancements in Software Defined Networks (SDN), Open Radio Access Network (O-RAN), and 5G technology have significantly expanded the capabilities of wireless networks, extending beyond mere data transmission. This progression has led to the emergence of Virtual Networks (VN) and Network Slicing, enabling industries to enhance their services and applications by establishing virtual networks that utilize shared physical infrastructure. Many works in the literature have considered optimizing the allocation of on-demand slices, assuming the absolute availability of resources and their accurate load. However, accurately allocating future network slices remains challenging due to the error in load prediction, diverse Key Performance Indicators (KPIs), resource price variations, and the potential for overor under-provisioning. This study presents a two-phase intelligent approach to address these challenges. The framework proactively predicts different slice loads while considering prediction errors in optimizing future slices with varied KPIs in a cost-efficient manner. Specifically, our method utilizes historical load data per service and employs AI-based forecasts for service load prediction. Subsequently, it employs a Deep Reinforcement Learning (DRL) agent on O-RAN's virtual Control Unit (vCU) and virtual Distributed unit (vDU) to correct errors in prediction and optimize the cost of slice allocation based on service KPI requirements, ultimately pre-allocating future network slices at reduced costs. Through experimental validation against various baselines and state-of-the-art solutions, we demonstrate the efficacy of our proposed solution, achieving a notable reduction (37-51%) in the average cost of allocated slices while inquiring about (1.5-7%

show abstract

On the Modeling of Reliability in Extreme Edge Computing Systems

Cited by 6 publications

References 12 publications

Multi-Objective Deep Reinforcement Learning for Efficient Workload Orchestration in Extreme Edge Computing

Multi-Objective Deep Reinforcement Learning for Efficient Workload Orchestration in Extreme Edge Computing

Incentive-Vacation Queueing in Extreme Edge Computing: An Analytical Reward-Based Framework

ECP: Error-Aware, Cost-Effective and Proactive Network Slicing Framework

Contact Info

Product

Resources

About