Optimizing apples lossless audio codec algorithm using NVIDIA CUDA

Abstract

As majority of the compression algorithms are implementations for CPU architecture, the primary focus of our work is to exploit the opportunities of GPU parallelism in audio compression. We present an implementation of Apples Lossless Audio Codec (ALAC) algorithm by using NVIDIA GPUs Compute Unified Device Architecture (CUDA) Framework. The core idea is to identify the areas where data parallelism can be applied and parallel programming model CUDA is used to execute the identified parallel components on Single Instruction Multiple Thread (SIMT) model of CUDA. The dataset is retrieved from European Broadcasting Union, Sound Quality Assessment Material (SQAM). Faster execution of the algorithm leads to execution time reduction when applied to audio coding for large audios. This paper also presents the reduction of power usage due to running the parallel components on GPU. Experimental results reveal that we achieve about 80-90% speedup through CUDA on the identified components over its CPU implementation while saving CPU power consumption.