상세정보
학술지
A layer-wise frequency scaling for a neural processing unit
- 발행일
- 202210
- 출처
- ETRI Journal, v.44 no.5, pp.849-858
- ISSN
- 1225-6463
- 출판사
- 한국전자통신연구원 (ETRI)
- 협약과제
- 22HS5600, 인공지능프로세서 전문연구실, 권영수
- 초록
- Dynamic voltage frequency scaling (DVFS) has been widely adopted for run-time power management of various processing units. In the case of neural processing units (NPUs), power management of neural network applications is required to adjust the frequency and voltage every layer to consider the power behavior and performance of each layer. Unfortunately, DVFS is inappropriate for layer-wise run-time power management of NPUs due to the long latency of voltage scaling compared with each layer execution time. Because the frequency scaling is fast enough to keep up with each layer, we propose a layer-wise dynamic frequency scaling (DFS) technique for an NPU. Our proposed DFS exploits the highest frequency under the power limit of an NPU for each layer. To determine the highest allowable frequency, we build a power model to predict the power consumption of an NPU based on a real measurement on the fabricated NPU. Our evaluation results show that our proposed DFS improves frame per second (FPS) by 33% and saves energy by 14% on average, compared with DVFS.
- KSP 제안 키워드
- Dynamic Frequency Scaling, Dynamic voltage and frequency scaling, Neural network applications, Neural processing, Power Consumption, Power Limit, Power behavior, Power model, Processing unit, Run time, Voltage scaling
본 저작물은 공공누리 제4유형 : 출처표시 + 상업적 이용금지 + 변경금지 조건에 따라 이용할 수 있습니다.
