지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학위논문
- 저자정보
- 지도교수
- 金亨信
- 발행연도
- 2013
- 저작권
- 충남대학교 논문은 저작권에 의해 보호받습니다.
이용수1
이 논문의 연구 히스토리 (2)
- 이 논문의 후속연구가 궁금하신가요?
- 연관 학술논문 또는 학술발표를 통해 보다 발전된 연구결과를 확인하실 수 있습니다.
- 이 논문의 연구 히스토리 확인하기
초록· 키워드
오류제보하기
During the past few years, smartphone technology has been rapidly developed, Especially, one of the notable features is that high-definition display on smartphone led to high-quality of graphic technology on mobile devices. As the display technology develops, smartphone users require more smooth UI and more realistic three-dimensional graphic games. It means increased graphic data processing on the mobile graphics processing unit(GPU).
Mobile GPU reduces increased burden on mobile CPU caused by computing for graphic data and accelerates the computation speed. Most of the recent smartphones are equipped with high-performance multi-core GPU.
How a multi-core mobile GPU can efficiently be utilized will be a critical issue for smartphone performance. However most of the existing researches are on a single-core mobile GPU, and researches for multi-core mobile GPU are rare.
In this paper, we analyzed job scheduling patterns and the efficiency of multi-core mobile GPU. For this research, we have chosen the Mali-400MP multi-core GPU from ARM which is commonly equipped with present smartphones. The DS-5 performance analyzer provided from ARM was used for profiling the Mali-400MP. In the profiling result, despite of the higher number of GPU cores, the total processing time required for certain graphics applications were increased. Also, we showed that when the GPU is processing for 3D game, substantial amount of overhead is caused by communicating between not only from the CPU to the GPU, but also within the GPUs.
Based on this result, we confirmed that more active research for multi-core mobile GPU should be performed to optimize present mobile GPUs. This paper can be the first step toward understanding mobile multi-core GPU.
Mobile GPU reduces increased burden on mobile CPU caused by computing for graphic data and accelerates the computation speed. Most of the recent smartphones are equipped with high-performance multi-core GPU.
How a multi-core mobile GPU can efficiently be utilized will be a critical issue for smartphone performance. However most of the existing researches are on a single-core mobile GPU, and researches for multi-core mobile GPU are rare.
In this paper, we analyzed job scheduling patterns and the efficiency of multi-core mobile GPU. For this research, we have chosen the Mali-400MP multi-core GPU from ARM which is commonly equipped with present smartphones. The DS-5 performance analyzer provided from ARM was used for profiling the Mali-400MP. In the profiling result, despite of the higher number of GPU cores, the total processing time required for certain graphics applications were increased. Also, we showed that when the GPU is processing for 3D game, substantial amount of overhead is caused by communicating between not only from the CPU to the GPU, but also within the GPUs.
Based on this result, we confirmed that more active research for multi-core mobile GPU should be performed to optimize present mobile GPUs. This paper can be the first step toward understanding mobile multi-core GPU.
목차
제 1 장 서 론 11.1 연구 배경 11.2 연구 소개 및 동기 21.3 관련 연구 41.4 논문의 기여도 8제 2 장 안드로이드 그래픽 102.1 OpenGL ES 102.1.1 Vertex shader 132.1.2 Fragment shader 142.2 안드로이드 3D 그래픽 14제 3 장 모바일 GPU 213.1 모바일 GPU 구조 213.2 Mali-400MP 멀티코어 GPU 273.2.1 Mali-400MP의 하드웨어 구조 283.2.2 Mali-400MP의 디바이스 드라이버 구조 36제 4 장 프로파일링 환경 구축 434.1 프로파일링 환경 434.2 프로파일링 도구 444.3 GPU 동작 분석을 위한 프로파일링 인터페이스 포팅 474.4 디바이스 드라이버 소스 레벨의 프로파일링 구현 50제 5 장 GPU의 작업 분배 및 효율성 분석 525.1 안드로이드 그래픽 프로그램에서의 GPU 코어 별 작업 분배 패턴 525.2 GPU 코어 수에 따른 그래픽 처리의 오버헤드 565.3 GPU 작업 분배의 효율성 분석 585.4 GPU shader 구조의 효율성 분석 59제 6 장 결론 616.1 결론 616.2 향후 계획 61참 고 문 헌 63ABSTRACT 66
최근 본 자료
댓글(0)
0