[논문]엣지 디바이스에서의 병렬 프로그래밍 모델 성능 비교 연구

남덕윤

doi:10.14372/iemek.2023.18.4.165

엣지 디바이스에서의 병렬 프로그래밍 모델 성능 비교 연구
A Performance Comparison of Parallel Programming Models on Edge Devices 원문보기

대한임베디드공학회논문지 = IEMEK Journal of embedded systems and applications, v.18 no.4, 2023년, pp.165 - 172

남덕윤 (Kyungpook National University)

Abstract ▼ AI-Helper

Heterogeneous computing is a technology that utilizes different types of processors to perform parallel processing. It maximizes task processing and energy efficiency by leveraging various computing resources such as CPUs, GPUs, and FPGAs. On the other hand, edge computing has developed with IoT and 5G technologies. It is a distributed computing that utilizes computing resources close to clients, thereby offloading the central server. It has evolved to intelligent edge computing combined with artificial intelligence. Intelligent edge computing enables total data processing, such as context awareness, prediction, control, and simple processing for the data collected on the edge. If heterogeneous computing can be successfully applied in the edge, it is expected to maximize job processing efficiency while minimizing dependence on the central server. In this paper, experiments were conducted to verify the feasibility of various parallel programming models on high-end and low-end edge devices by using benchmark applications. We analyzed the performance of five parallel programming models on the Raspberry Pi 4 and Jetson Orin Nano as low-end and high-end devices, respectively. In the experiment, OpenACC showed the best performance on the low-end edge device and OpenSYCL on the high-end device due to the stability and optimization of system libraries.

주제어

표/그림 (9)

그림 그림 1. 이기종 엣지 서버와 디바이스 예시 Fig. 1. Example of heterogeneous edge server and devices
그림 그림 2. 실험용 디바이스: 라즈베리 파이4와 젯슨 오린 나노 Fig. 2. Experimental devices: Raspberry Pi 4 and Jetson Orin Nano
표 표 1. 실험 디바이스 주요 사양 Table 1. Main specifications of experimental devices
표 표 2. 실험 디바이스별 소프트웨어 목록 Table 2. List of software on experimental devices
표 표 3. 디바이스 자원별 프로그래밍 모델 실행 여부 Table 3. Availability of programming model by device resource
표 표 4. 저사양 디바이스에서의 대역폭 측정 결과 (MB/sec) Table 4. Bandwidth results measured on a low-end device
표 표 5. 저사양 디바이스에서의 병렬 프로그래밍 모델별 주요 시스템 콜 현황 Table 5. Statistics of system calls by parallel programming frameworks on a low-end device
표 표 6. 고사양 디바이스에서의 대역폭 측정 결과 (MB/sec) Table 6. Bandwidth results measured on a high-end device
표 표 7. 고사양 디바이스에서의 병렬 프로그래밍 모델별 주요 시스템 콜 현황 Table 7. Statistics of system calls by parallel programming frameworks on a high-end device

참고문헌 (25)

V. G. Cerf, "On Heterogeneous Computing,"？Communication of the ACM, Vol. 64, No. 21, pp. 9, 2021.
OpenCL, https://www.khronos.org/opencl/
S. Mendez, "Edge Computing Systems with？Kubernetes," Packt Publishing, 2022.
J. Diaz, C. Munoz-Caro, A. Nino, "A Survey of Parallel？Programming Models and Tools in the Multi and？Many-Core Era," IEEE Transactions on Parallel and？Distributed Systems, Vol. 23, No. 8, pp. 1369-1386, 2012.

상세보기
CUDA, https://developer.nvidia.com/cuda-toolkit
SYCL, https://www.khronos.org/sycl/
T. Deakin, J. Price, M. Martineau, S. McIntosh-Smith,？"Evaluating Attainable Memory Bandwidth of Parallel？Programming Models Via BabelStream," International？Journal of Computational Science and Engineering, Vol.？17, No. 3, pp. 247-262, 2018.

상세보기
Raspberry Pi, https://www.raspberrypi.com/products/raspberry-pi-4-model-b/
Jetson Orin Nano, https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin/
A. Alpay, V. Heuveline, "One Pass to Bind Them: The？First Single-Pass SYCL Compiler with Unified Code？Representation Across Backends," in Proceedings of the？2023 International Workshop on OpenCL, Article 7, 2023.
OpenMP, https://www.openmp.org/
OpenACC, https://www.openacc.org/
D. Angus, S. Georgiev, H. A. Gonzalez, J. Riordan, P.？Keir, M. Goli, "Porting SYCL Accelerated Neural？Network Frameworks to Edge Devices," in Proceedings？of the 2023 International Workshop on OpenCL, Article？No. 4, 2023.
J. Y. Park, J. H. Hong, K. S. Chung "Parallel LDPC？Decoder for CMMB on CPU and GPU Using OpenCL,"？IEMEK J. Embed. Sys. Appl., Vol. 11, No. 6, pp.？325-334, 2016 (in Korean).
S. Memeti, L. Li, S. Pllana, J. Kolodziej, C. Kessler,？"Benchmarking OpenCL, OpenACC, OpenMP, and？CUDA: Programming Productivity, Performance, and？Energy Consumption," in Proceedings of the 2017？Workshop on Adaptive Resource Management and？Scheduling for Cloud Computing, pp. 1-7, 2017.
Codeplay ComputeCpp, https://developer.codeplay.com/products/computecpp/ce/home
Intel oneAPI DPC++, https://www.intel.com/content/www/us/en/developer/tools/oneapi/dpc-compiler.html
OpenSYCL, https://github.com/OpenSYCL/OpenSYCL
STREAM Benchmark, https://www.cs.virginia.edu/stream/
VC4CL: OpenCL for VideoCore IV GPU, https://github.com/doe300/VC4CL
strace: Linux syscall tracer, https://strace.io/
JetPack SDK, https://developer.nvidia.com/embedded/jetpack
Portable Computing Language, http://portablecl.org/
NVIDIA HPC SDK, https://developer.nvidia.com/hpc-sdk
C. Feng, P. Han X. Zhang, B. Yang, Y. Liu, L. Guo,？"Computation Offloading in Mobile Edge Computing？Networks: A Survey," Journal of Network and？Computer Applications, Vol. 202, 103366, 2022.

상세보기

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증