nach oben

The Journal of Supercomputing

17.05.2024

DGCQN: a RL and GCN combined method for DAG scheduling in edge computing

verfasst von: Bin Qin, Qinyang Lei, Xin Wang

Erschienen in: The Journal of Supercomputing

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config

KI-gestützte Suche

Aus

Abstract

Edge computing is an emerging paradigm that enables low-latency and high-performance computing at the network edge. However, effectively scheduling complex and interdependent tasks on heterogeneous and dynamic edge computing nodes presents a significant challenge in meeting users' real-time response requirements. To solve this problem, a DGCQN scheduling network that leverages reinforcement learning and graph convolutional neural networks to learn an optimal scheduling strategy is proposed in this paper. The proposed method embeds the graph structure of Directed Acyclic Graph (DAG) tasks and node information of Kubernetes (K8s) clusters into a Q value function, guiding the DQN network in selecting the best action at each step. The method is evaluated across various DAG tasks and edge computing scenarios. Compared with HEFT, DQN, and GOSU, the task completion time of the proposed method is reduced by about 20, 10, and 1.5%, respectively. The results demonstrate the effectiveness of the proposed method.

Graphic abstract

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 340 Zeitschriften

aus folgenden Fachgebieten:

Bauwesen + Immobilien
Business IT + Informatik
Finance + Banking
Management + Führung
Marketing + Vertrieb
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

über 67.000 Bücher
über 390 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Maschinenbau + Werkstoffe

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

über 102.000 Bücher
über 537 Zeitschriften

aus folgenden Fachgebieten:

Automobil + Motoren
Bauwesen + Immobilien
Business IT + Informatik
Elektrotechnik + Elektronik
Energie + Nachhaltigkeit
Finance + Banking
Management + Führung
Marketing + Vertrieb
Maschinenbau + Werkstoffe
Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Jetzt informieren

Shi W, Cao J, Zhang Q et al (2016) Edge computing: vision and challenges. IEEE Internet Things J 3(5):637–646. https://doi.org/10.1109/JIOT.2016.2579198CrossRef

Sufyan F, Banerjee A (2023) Computation offloading for smart devices in fog-cloud queuing system. IETE J Res 69(3):1509–1521. https://doi.org/10.1080/03772063.2020.1870876CrossRef

Sufyan F, Banerjee A (2020) Computation offloading for distributed mobile edge computing network: a multiobjective approach. IEEE Access 8:149915–149930. https://doi.org/10.1109/ACCESS.2020.3016046CrossRef

Chishti MS, Sufyan F, Banerjee A (2021) Decentralized on-chain data access via smart contracts in Ethereum blockchain. IEEE Trans Netw Serv Manage 19(1):174–187. https://doi.org/10.1109/TNSM.2021.3120912CrossRef

Sufyan F, Banerjee A (2019) Comparative analysis of network libraries for offloading efficiency in mobile cloud environment. Int J Adv Comput Sci Appl, 10(2)

Chiang YH, Zhang T, Ji Y (2019) Joint cotask-aware offloading and scheduling in mobile edge computing systems. IEEE Access 7:105008–105018. https://doi.org/10.1109/ACCESS.2019.2931336CrossRef

Li C, Tang J, Tang H et al (2019) Collaborative cache allocation and task scheduling for data-intensive applications in edge computing environment. Futur Gener Comput Syst 95:249–264. https://doi.org/10.1016/j.future.2019.01.007CrossRef

Khan PW, Abbas K, Shaiba H et al (2020) Energy efficient computation offloading mechanism in multi-server mobile edge computing—an integer linear optimization approach. Electronics 9(6):1010. https://doi.org/10.3390/electronics9061010CrossRef

Qu Y, Dai H, Wang H et al (2021) Service provisioning for UAV-enabled mobile edge computing. IEEE J Sel Areas Commun 39(11):3287–3305. https://doi.org/10.1109/JSAC.2021.3088660CrossRef

10.

Xiuguo Z, Yufei L, Zhiying C et al (2021) An efficient service migration model based on improved genetic algorithm in mobile edge computing environment. 革新的コンピューティング・情報・制御に関する国際誌 17(04):1401

11.

Sheng S, Chen P, Chen Z et al (2021) Deep reinforcement learning-based task scheduling in iot edge computing. Sensors 21(5):1666. https://doi.org/10.3390/s21051666MathSciNetCrossRef

12.

Jie Y, Tang X, Choo KKR et al (2018) Online task scheduling for edge computing based on repeated Stackelberg game. J Parall Distribut Comput 122:159–172. https://doi.org/10.1016/j.jpdc.2018.07.019CrossRef

13.

Gezer V, Wagner A (2021) Real-time edge framework (RTEF): task scheduling and realisation. J Intell Manuf 2021:1–17. https://doi.org/10.1007/s10845-021-01760-9CrossRef

14.

Cai L, Wei X, Xing C et al (2021) Failure-resilient DAG task scheduling in edge computing. Comput Netw 198:108361. https://doi.org/10.1016/j.comnet.2021.108361CrossRef

15.

Zhan W, Luo C, Wang J et al (2020) Deep-reinforcement-learning-based offloading scheduling for vehicular edge computing. IEEE Internet Things J 7(6):5449–5465. https://doi.org/10.1109/JIOT.2020.2978830CrossRef

16.

Zhu A, Wen Y (2021) Computing offloading strategy using improved genetic algorithm in mobile edge computing system. J Grid Comput 19(3):38. https://doi.org/10.1007/s10723-021-09578-8MathSciNetCrossRef

17.

Zhang Y, Wang Y, Tang X et al (2018) Energy-efficient task scheduling on heterogeneous computing systems by linear programming. Concurrency Comput Pract Exp 30(19):e4731. https://doi.org/10.1002/cpe.4731CrossRef

18.

Leng L, Li J, Shi H et al (2021) Graph convolutional network-based reinforcement learning for tasks offloading in multi-access edge computing. Multimedia Tools Appl 80(19):29163–29175. https://doi.org/10.1007/s11042-021-11130-5CrossRef

19.

Tang Z, Jia W, Zhou X et al (2020) Representation and reinforcement learning for task scheduling in edge computing. IEEE Trans Big Data 8(3):795–808. https://doi.org/10.1109/TBDATA.2020.2990558CrossRef

20.

Lee H, Cho S, Jang Y et al (2021) A global DAG task scheduler using deep reinforcement learning and graph convolution network. IEEE Access 9:158548–158561.https://doi.org/10.1109/ACCESS.2021.3130407

21.

Jing X, Yao X, Liu M et al (2024) Multi-agent reinforcement learning based on graph convolutional network for flexible job shop scheduling. J Intell Manuf 35(1):75–93. https://doi.org/10.1007/s10845-022-02037-5CrossRef

22.

Lin Z, Li C, Tian L et al (2022) A scheduling algorithm based on reinforcement learning for heterogeneous environments. Appl Soft Comput 130:109707. https://doi.org/10.1016/j.asoc.2022.109707CrossRef

23.

Song Y, Li C, Tian L et al (2023) A reinforcement learning based job scheduling algorithm for heterogeneous computing environment. Comput Electr Eng 107:108653. https://doi.org/10.1016/j.compeleceng.2023.108653CrossRef

24.

Topcuoglu H, Hariri S, Wu M Y (2002) Performance-effective and low-complexity task scheduling for heterogeneous computing. IEEE Trans Parall Distribut Syst 13(3):260–274.https://doi.org/10.1109/71.993206

25.

Muddinagiri R, Ambavane S, Bayas S (2019) Self-hosted Kubernetes: deploying Docker containers locally with Minikube. In: 2019 IEEE International Conference on Innovative Trends and Advances in Engineering and Technology (ICITAET): 239–243.https://doi.org/10.1109/ICITAET47105.2019.9170208

26.

Kubernetes (2022) Production-Grade Container Orchestration. Available online: https://kubernetes.io/ (accessed on 15 July 2022)

27.

Mondal SK, Pan R, Kabir HMD et al (2022) Kubernetes in IT administration and serverless computing: an empirical study and research challenges. J Supercomput 2022:1–51. https://doi.org/10.1007/s11227-021-03982-3CrossRef

28.

Khalil E, Dai H, Zhang Y et al. (2017) Learning combinatorial optimization algorithms over graphs. Adv Neural Inf Process Syst, 30.

Titel: DGCQN: a RL and GCN combined method for DAG scheduling in edge computing
verfasst von: Bin Qin
Qinyang Lei
Xin Wang
Publikationsdatum: 17.05.2024
Verlag: Springer US
Erschienen in: The Journal of Supercomputing
Print ISSN: 0920-8542
Elektronische ISSN: 1573-0484
DOI: https://doi.org/10.1007/s11227-024-06140-7

Springer Professional

Abstract

Graphic abstract

Bitte loggen Sie sich ein, um Zugang zu Ihrer Lizenz zu erhalten.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft"

Springer Professional "Technik"

Springer Professional "Wirtschaft+Technik"

Premium Partner