Deep reinforcement learning-based microservice selection in mobile edge computing

Guo, Feiyan; Tang, Bing; Tang, Mingdong; Liang, Wei

doi:10.1007/s10586-022-03661-9

Cited by 12 publications

(5 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gupta et al [28] implement enhanced algorithms (e.g., Max-Min and Greedy) for load balancing in cloud environments. Qiu et al [29], Guo et al [30], Li [31], and Ali [32] adopt machine learning models (e.g., reinforcement learning and transfer learning) as the main algorithms for service deployment, microservice selection, optimizing the delay, and reducing deployment cost with a fixed service set. However, due to the issue of model retraining, theses algorithms may not be suitable for a microservices system with new services within short execution time.…”

Section: B Cloud Environments 1) Single-cloud Scenariomentioning

confidence: 99%

“…Architecture Management Objectives Algorithm/Platform [19] Single cloud Allocation Resource PSO [20] Single cloud Allocation Energy consumption SA,tabu search [27] Single cloud Scheduling Performance ACO [26] Single cloud Scheduling Performance, association, clustering Own algorithm [28] Single cloud Allocation Resource Max-Min, Greedy [29] Single cloud Allocation Resource Machine learning [30] Single cloud Scheduling Microservice selection Machine learning [31] Single cloud Service deployment Delay and deployment cost Machine learning [22] Multiple clouds Allocation Service cost, latency, availability NSGA-II [23] Multiple clouds Allocation Load balancing NSGA-II [24] Multiple clouds Allocation Availability, energy consumption NSGA-II [33] Multiple clouds Placement CPU performance, microservice interaction Greedy algorithm [34] Multiple clouds Scheduling Resource, availability, costs GA [35] Multiple clouds Traffic Availability, reliability Grafana and Prometheus [36] Multiple clouds Allocation Resource, cost Own algorithm [37] Heterogeneous cloud Scheduling Performance, energy consumption Own algorithm [38] Heterogeneous cloud Scheduling Execution times, costs NSGA-II [39] Heterogeneous cloud Allocation Resource Own framework This work…”

Section: Referencementioning

confidence: 99%

See 1 more Smart Citation

Optimal Resource Allocation Using Genetic Algorithm in Container-Based Heterogeneous Cloud

Chen,

Wen

2024

IEEE Access

View full text Add to dashboard Cite

This paper tackles the complex problem of optimizing resource configuration for microservice management in heterogeneous cloud environments. To address this challenge, an enhanced framework, the multi-objective microservice allocation (MOMA) algorithm, is developed to formulate the efficient resource management of cloud microservice resources as a constrained optimization problem, guided by resource utilization and network communication overhead, which are two important factors in microservice resource allocation. The proposed framework simplifies the deployment of cloud services and streamlines workload monitoring and analysis within a diverse cloud system. A comprehensive comparison is made between the effectiveness of the proposed algorithm and existing algorithms on real-world datasets, with a focus on resource balancing, network overhead, and network reliability. Experimental results reveal that the proposed algorithm significantly enhances resource utilization, reduces network transmission overhead, and improves reliability.

show abstract

Section: B Cloud Environments 1) Single-cloud Scenariomentioning

confidence: 99%

Section: Referencementioning

confidence: 99%

Optimal Resource Allocation Using Genetic Algorithm in Container-Based Heterogeneous Cloud

Chen,

Wen

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Consequently, many scholars focus on these heterogeneous servers as subjects for research, employing methods such as linear programming to address the challenges of service deployment [8]. Microservices and container technology have also been widely applied in edge environments [9]. As user demands grow more complex, the architecture of applications becomes more intricate, and monolithic applications often fail to meet user needs in many respects [5].…”

Section: Introductionmentioning

confidence: 99%

Microservice Deployment in Cloud-Edge Environment using Enhanced Global Search Grey Wolf Optimizer-Greedy Algorithm

Wang,

Zhang,

et al. 2024

Preprint

View full text Add to dashboard Cite

The rapid advancement of edge-cloud technologies has made service deployment increasingly crucial. Additionally, benefiting from the reusability of services, complex applications are subdivided into different microservices. Given the constraints of limited resources, heterogeneous servers, and the geographical diversity of users, how to reasonably deploy microservices becomes a significant challenge. In this paper, we propose a microservice deployment model aimed at minimizing users' latency and maximizing edge providers' profits. The model is divided into different scenarios, each with varying trends in user request categories. To seek microservice deployment strategies, we introduce an Enhanced Global Search Grey Wolf Optimizer-Greedy (EGSGWO-G) algorithm designed for microservice deployment-offloading frameworks. This algorithm leverages EGSGWO to search for deployment strategies and evaluates them using greedy service offloading algorithm. Finally, extensive experiments demonstrate that the EGSGWO-G algorithm improves convergence speed by 31.78%, reduces latency by 12.64%, and increases provider profits by 1.30% compared to GWO-G.

show abstract

“…[7][8][9] Especially, under the hybrid cloud-edge collaborative architecture, how to effectively manage containers is a challenging issue. 10,11 Some works have been done to adapt Kubernetes to edge computing environment, such as the open-source projects, KubeEdge ‡ , EdgeGallery § , SuperEdge ¶ , OpenYurt # and so forth.…”

Section: Introductionmentioning

confidence: 99%

“…How to adapt Docker and container technology to the edge has become a current hot research issue, which has emerged the concept of edge cloud and cloud native for edge computing 7‐9 . Especially, under the hybrid cloud‐edge collaborative architecture, how to effectively manage containers is a challenging issue 10,11 . Some works have been done to adapt Kubernetes to edge computing environment, such as the open‐source projects, KubeEdge, EdgeGallery, SuperEdge, OpenYurtand so forth.…”

Section: Introductionmentioning

confidence: 99%

Container‐based data‐intensive application scheduling in hybrid cloud‐edge collaborative environment

2023

Self Cite

View full text Add to dashboard Cite

Container virtualization technology represented by Docker has been widely used in the industry due to its advantages of lightweight, fast deployment, and easy portability. It can bring convenience to system deployment, operation, and maintenance. This article considers the implementation of AI‐based IoT applications based on Docker and container technology in a cloud‐edge collaborative environment. Three typical data‐intensive applications, preprocessing, training, and inference in machine learning, are deployed using Docker containers. For these three types of tasks, this article proposes a container‐based data‐intensive application scheduling framework. Using an attribute‐based data‐driven method, a new scheduling approach considering multiple weighting factors is proposed, which is called DICS‐OPT. The cloud nodes and edge nodes are scored and sorted uniformly. The container is scheduled to the node with the highest score, and then the task data is transmitted to the node to execute the task. NSGA‐III‐based genetic algorithm has been proposed to search for the optimal weighting factors. This article introduces the implementation of the prototype system and builds a cloud‐edge collaborative testbed consisting of Raspberry Pis and PCs. The performance evaluation results indicates that the proposed scheduling approach outperforms existing approaches. Compared with existing approaches, DICS‐OPT improves the average edge resource utilization by 10.05% to 69.04%, and saves the cloud resource cost by 16.02% to 36.68%.

show abstract

Deep reinforcement learning-based microservice selection in mobile edge computing

Cited by 12 publications

References 41 publications

Optimal Resource Allocation Using Genetic Algorithm in Container-Based Heterogeneous Cloud

Optimal Resource Allocation Using Genetic Algorithm in Container-Based Heterogeneous Cloud

Microservice Deployment in Cloud-Edge Environment using Enhanced Global Search Grey Wolf Optimizer-Greedy Algorithm

Container‐based data‐intensive application scheduling in hybrid cloud‐edge collaborative environment

Contact Info

Product

Resources

About