IEMEK Journal of Embedded Systems and Applications (대한임베디드공학회논문지)

Institute of Embedded Engineering of Korea (대한임베디드공학회)
권호 표지
DOI QR코드

DOI QR Code

Kubernetes Scheduler Framework Implementation with Realtime Resource Monitoring

실시간 자원 모니터링을 활용한 쿠버네티스 스케줄러 프레임워크 구현

  • Received : 2019.11.13
  • Accepted : 2020.06.05
  • Published : 2020.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Kubernetes is one of the most widely used tool in cloud computing environments. It m anages POD as a basic operational unit, providing a isolated environments through container tech nology. Basically, its scheduler properly allocates POD, considering the utilizations of CPUs, mem ories and volumes on hosts. In our work, we develop a customized scheduling framework additio nally considering network resources. In the framework, we can monitor the dynamic variations of resources and make it possible to utilize the resources for the scheduler. This framework offers not only observing necessary information but also visualizing the data to scheduler for providing convenience.

Keywords

References

  1. Kim Dongmin, Hanif Muhammad, Eunsam Kim, Sumi Helal, Choonhaw Lee, "TOSCA-Based, Federation-Aware Cloud Orchestration for Kubernetes Container Platform," Applied Sciences, Vol. 9, pp. 191, 2019. https://doi.org/10.3390/app9010191
  2. https://docker.com/resources/what-container/.
  3. https://kubernetes.io.
  4. Chia-Chen Chang, Shun-Ren Yang, En-Hau Yeh, Phone Lin, Jeu-Yih Jeng, "A Kubernetes-based Monitoring Platform for Dynamic Cloud Resource Provisioning," GLOBECOM 2017-2017 IEEE Global Communications Conference, pp. 1-6, 2017.
  5. Victor Medel, Omer Rana, Jose Angel Banares, Unai Arronategui, "Modelling Performance & Resource Management in Kubernetes," Proceedings of the 9th International Conference on Utility and Cloud Computing, pp. 257-262, 2016.
  6. https://kubernetes.io/docs/reference/scheduling/policies.
  7. Elana Hashman. "Operating Within Normal Parameters: Monitoring Kubernetes," Proceedings of USENIX SRECON'19, 2019.