97. approximation for energy efficient computing
Name: Mohsen Imani
Grad Year: 2019
The Internet of Things (IoT) dramatically increases the amount of data to be processed for many applications including multimedia. Unlike traditional computing environment, the workload of IoT significantly varies overtime. Thus, an efficient runtime profiling is required to extract highly frequent computations and pre-store them for memory-based computing. In this paper, we propose a technique which utilizes using a low-cost lookup table to enable approximation. Evaluating the proposed design on the recent GPU architecture shows that the proposed design can achieve up to 60% energy savings and 30% speed up for image processing applications with an acceptable quality of service.
Industry Application Area(s)
Electronics/Photonics | Internet, Networking, Systems | Semiconductor