Characterization and Throttling-Based Mitigation of Memory Interference for Heterogeneous Smartphones

Davesh Shingari, Akhil Arunkumar, Carole-Jean Wu
2015 2015 IEEE International Symposium on Workload Characterization  
The availability of a wide range of general-purpose as well as accelerator cores on modern smartphones means that a significant number of applications can be executed on a smartphone simultaneously, resulting in an ever increasing demand on the memory subsystem. While the increased computation capability is intended for improving user experience, memory requests from each concurrent application exhibit unique memory access patterns as well as specific timing constraints. If not considered, this
more » ... could lead to significant memory contention and result in lowered user experience. In this paper, we design experiments to analyze the performance degradation caused by the interference at the memory subsystem for a broad range of commonly-used smartphone applications. The characterization studies are performed on a real smartphone device -Google Nexus5 -running an Android operating system. Our results show that user-centric smartphone applications, such as web browsing and media player, suffer upto 34% and 21% performance degradation, respectively, from shared resource contention at the application processor's last-level cache, the communication fabric, and the main memory. Taking a step further, we demonstrate the feasibility and effectiveness of a frequency throttling-based memory interference mitigation technique. At the expense of performance degradation of interfering applications, frequency throttling is an effective technique for mitigating memory interference, leading to better QoS and user experience, for user-centric applications. programmed use case consists of file downloading via the Wi-Fi/LTE antenna, music playback via a specific accelerator, and web browsing on the application cores, generating a heterogeneous combination of memory requests at the main memory. Furthermore, as GPUs are becoming more and more programmable, general-purpose computations are increasingly offloaded to mobile GPUs, e.g., [2] , to achieve higher performance and improved energy efficiency. Similarly, more and more accelerators have been added to modern smartphone SoCs to execute special functions as energy efficiently as possible, e.g., Qualcomm's programmable digital signal processor (DSP) released in 2013, and many more other low-power accelerators in the years to come [3], [4] .
doi:10.1109/iiswc.2015.9 dblp:conf/iiswc/ShingariAW15 fatcat:zb53b24z2baizbubkhksj5ncwe