Approximating Memory-bound Applications on Mobile GPUs

Approximate computing techniques are often used to improve the performance of applications that can tolerate some amount of impurity in the calculations or data. In the context of embedded and mobile systems, a broad number of applications have exploited approximation techniques to improve performance and overcome the limited capabilities of the hardware. On such systems, even small performance improvements can be sufficient to meet scheduled requirements such as hard real-time deadlines. We study the approximation of memory-bound applications on mobile GPUs using kernel perforation, an approximation technique that exploits the availability of fast GPU local memory to provide high performance with more accurate results. Using this approximation technique, we approximated six applications and evaluated them on two mobile GPU architectures with very different memory layouts: a Qualcomm Adreno 506 and an ARM Mali T860 MP2. Results show that, even when the local memory is not mapped to dedicated fast memory in hardware, kernel perforation is still capable of $1.25\times$ speedup because of improved memory layout and caching effects. Mobile GPUs with local memory show a speedup of up to $1.38\times$.