Optimizing Heterogeneous Task Allocation for Edge Compute Micro Clusters Using PSO Metaheuristic

Alhaizaey, Yousef; Singer, Jeremy; Michala, Anna Lito

doi:10.1109/fmec57183.2022.10062755

Cited by 6 publications

(7 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The study in [20] extends previous research on task management by proposing a linear-based model coupled with a metaheuristic Particle Swarm Optimization (PSO) approach. The study aims to optimize the makespan time and the task allocation overhead for heterogeneous edge workload management [20]. This study complements and informs the proposed research by providing insights into the scalability and efficiency of various allocation mechanisms in micro-cluster IoT edge computing.…”

Section: Related Workmentioning

confidence: 69%

“…The task management in IoT edge computing, mainly the edge micro-cluster platforms, is vital for efficiently tackling IoT applications that demand Quality of Service (QoS) [20]. The study in [20] extends previous research on task management by proposing a linear-based model coupled with a metaheuristic Particle Swarm Optimization (PSO) approach. The study aims to optimize the makespan time and the task allocation overhead for heterogeneous edge workload management [20].…”

Section: Related Workmentioning

confidence: 89%

“…The authors of [19] introduced an evolutionary mechanism for IoT edge task clustering and scheduling to optimize the cost and energy consumption. Similarly, the study in [20] developed a Particle Swarm Optimization (PSO) mechanism for edge device task distribution. While important, these related works do not address the unique constraints and real-time issues present in IoT edge task scheduling, a gap the proposed DE mechanism aims to fill.…”

Section: Related Workmentioning

confidence: 99%

“…The problem under this study involves scheduling tasks on a set of IoT edge computing resources to minimize the total execution time [20,21]. The IoT edge tasks are divided into clusters based on their similarities, and each cluster is assigned to a resource for execution [23,36].…”

Section: Problem Formulationmentioning

confidence: 99%

See 3 more Smart Citations

An Evolutionary Algorithm for Task Clustering and Scheduling in IoT Edge Computing

Yousif,

Bashir,

Ali

2024

Mathematics

View full text Add to dashboard Cite

The Internet of Things (IoT) edge is an emerging technology of sensors and devices that communicate real-time data to a network. IoT edge computing was introduced to handle the latency concerns related to cloud computing data management, as the data are processed closer to their point of origin. Clustering and scheduling tasks on IoT edge computing are considered a challenging problem due to the diverse nature of task and resource characteristics. Metaheuristics and optimization methods are widely used in IoT edge task clustering and scheduling. This paper introduced a new task clustering and scheduling mechanism using differential evolution optimization on IoT edge computing. The proposed mechanism aims to optimize task clustering and scheduling to find optimal execution times for submitted tasks. The proposed mechanism for task clustering is based on the degree of similarity of task characteristics. The proposed mechanisms use an evolutionary mechanism to distribute system tasks across suitable IoT edge resources. The clustering tasks process categorizes tasks with similar requirements and then maps them to appropriate resources. To evaluate the proposed differential evolution mechanism for IoT edge task clustering and scheduling, this study conducted several simulation experiments against two established mechanisms: the Firefly Algorithm (FA) and Particle Swarm Optimization (PSO). The simulation configuration was carefully created to mimic real-world IoT edge computing settings to ensure the proposed mechanism’s applicability and the simulation results’ relevance. In the heavyweight workload scenario, the proposed DE mechanism started with an execution time of 916.61 milliseconds, compared to FA’s 1092 milliseconds and PSO’s 1026.09 milliseconds. By the 50th iteration, the proposed DE mechanism had reduced its execution time significantly to around 821.27 milliseconds, whereas FA and PSO showed lesser improvements, with FA at approximately 1053.06 milliseconds and PSO stabilizing at 956.12 milliseconds. The simulation results revealed that the proposed differential evolution mechanism for edge task clustering and scheduling outperforms FA and PSO regarding system efficiency and stability, significantly reducing execution time and having minimal variation across simulation iterations.

show abstract

Section: Related Workmentioning

confidence: 69%

Section: Related Workmentioning

confidence: 89%

Section: Related Workmentioning

confidence: 99%

Section: Problem Formulationmentioning

confidence: 99%

See 2 more Smart Citations

An Evolutionary Algorithm for Task Clustering and Scheduling in IoT Edge Computing

Yousif,

Bashir,

Ali

2024

Mathematics

View full text Add to dashboard Cite

show abstract

“…Their performance could degrade in noisy environments, and they were sometimes insufficient in capturing the nuances of different speaking styles and accents. Deep learning, especially Deep Neural Networks (DNN), introduced by (Akhtarshenas, Vahedifar, Ayoobi, Maham, & Alizadeh, 2023;Alhaizaey, 2023;Y. Wang et al, 2023), has significantly impacted speaker recognition.…”

Section: Literature Reviewmentioning

confidence: 99%

Evaluating the Efficacy of Traditional Machine Learning Models in Speaker Recognition: A Comparative Study Using the LibriSpeech Dataset

Airlangga

2024

Brilliance

View full text Add to dashboard Cite

The efficacy of machine learning models in speaker recognition tasks is critical for advancements in security systems, biometric authentication, and personalized user interfaces. This study provides a comparative analysis of three prominent machine learning models: Naive Bayes, Logistic Regression, and Gradient Boosting, using the LibriSpeech test-clean dataset—a corpus of read English speech from audiobooks designed for training and evaluating speech recognition systems. Mel-Frequency Cepstral Coefficients (MFCCs) were extracted as features from the audio samples to represent the power spectrum of the speakers’ voices. The models were evaluated based on precision, recall, F1-score, and accuracy to determine their performance in correctly identifying speakers. Results indicate that Logistic Regression outperformed the other models, achieving nearly perfect scores across all metrics, suggesting its superior capability for linear classification in high-dimensional spaces. Naive Bayes also demonstrated high efficiency and robustness, despite the inherent assumption of feature independence, while Gradient Boosting showed slightly lower performance, potentially due to model complexity and overfitting. The study underscores the potential of simpler machine learning models to achieve high accuracy in speaker recognition tasks, particularly where computational resources are limited. However, limitations such as the controlled nature of the dataset and the focus on a single feature type were noted, with recommendations for future research to include more diverse environmental conditions and feature sets.

show abstract

Metaheuristic task offloading approaches for minimization of energy consumption on edge computing: a systematic review

Latip,

Aminu,

Hanafi

et al. 2024

Discov Internet Things

View full text Add to dashboard Cite

Optimizing Heterogeneous Task Allocation for Edge Compute Micro Clusters Using PSO Metaheuristic

Cited by 6 publications

References 21 publications

An Evolutionary Algorithm for Task Clustering and Scheduling in IoT Edge Computing

An Evolutionary Algorithm for Task Clustering and Scheduling in IoT Edge Computing

Evaluating the Efficacy of Traditional Machine Learning Models in Speaker Recognition: A Comparative Study Using the LibriSpeech Dataset

Metaheuristic task offloading approaches for minimization of energy consumption on edge computing: a systematic review

Contact Info

Product

Resources

About