An empirical study of Mobile Code Offloading in Unpredictable Environments

Mobile code offloading is a well-known technique for enhancing the capabilities of mobile platforms by transparently leveraging resources to cloud. Although this has been studied years, little empirical evidence exists demonstrate its alleged benefits in terms performance real-life situations. All studies conducted on topic have so far relegated controlled environments laboratory settings. As such, there no how and well performs scenarios, where network unreliability norm. In work, we present first study an Android application integrated with framework being tested wild. We distributed that contains set benchmarks APK format deployed it wide gamut devices which had physical access. carefully detail methodology infrastructure used monitor benchmarks’ 18 devices. Overall, our results show accuracy decision-making engine heavily affected couple factors, mainly diagnosis connection type. Therefore, determining whether or not more convenient execute given task cloud difficult task. summarize five lessons learned performing experiment believe should be considered future experiments area.