HeteroAccel: 엣지 컴퓨팅 환경에서의 다양한 영상 추론을 위한 쿠버네티스 기반의 이종 연산·가속기 자원 관리 시스템
Kubernetes-based Heterogeneous Computational and Accelerator Resource Management System for Various Image Inferences in Edge Computing Environments
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Edge Computing enables image-based inference in close proximity to end users and real-world objects. However, since edge servers have limited computational and accelerator resources, efficient resource management is essential. In this paper, we present HeteroAccel system that performs optimal scheduling in Kubernetes platform based on available node and accelerator information for various inference requests. Our experiments showed 25.3% improvement in overall inference performance over the default scheduling scheme in edge computing environment in which four types of inference services are requested.
Abstract
Edge Computing enables image-based inference in close proximity to end users and real-world objects. However, since edge servers have limited computational and accelerator resources, efficient resource management is essential. In this paper, we present HeteroAccel system that performs optimal scheduling in Kubernetes platform based on available node and accelerator information for various inference requests. Our experiments showed 25.3% improvement in overall inference performance over the default scheduling scheme in edge computing environment in which four types of inference services are requested.
참고문헌 (12)
- J. Ren, "A Survey on End-Edge-Cloud Orchestrated Network Computing Paradigms : Transparent Computing, Mobile Edge Computing, Fog Computing, and Cloudlet", ACM Computing Surveys, 52(6), pp.1~36 (2020)
- X. Wang, "Convergence of Edge Computing and deep Learning : A Comprehensive Survey", IEEE Communications Surveys & Tutorials, 22(2), pp.869~904 (2020)
- "Karbon 700"[URL]
- "MK400-70"[URL]
- 김세진, "GPU 컨테이너 동시 실행에 따른 응용의 간섭 측정 프레임워크 설계", KNOM Review, 23(1), pp.43~50 (2020), 한국통신학회[DOI]
- "2020", ~
- G. Cho, "Hybrid Resource Scheduling Scheme for Video Surveillance in GPU-FPGA Accelerated Edge System", Masters
- Xing Liu, "Collaborative Edge Computing With FPGA-Based CNN Accelerators for Energy-Efficient and Time-Aware Face Tracking System", IEEE Transactions on Computational Social Systems, pp.1~15 (2021), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- "OpenVINO"[URL]
- "Kubernetes"[URL]
- Z. Zhong, "A Cost-Efficient Container Orchestration Strategy in Kubernetes-Based Cloud Computing Infrastructures with Heterogeneous Resources", ACM Trans. on Internet Technology, 20(2), pp.1~24 (2021)
- "Prometheus"[URL]