Study On Optimization Of Mobile Storage System Base On Swapping Method

Smartphones are not phones any more.By equipping high performance mobile processors and large main memory,smartphones are now integrated more applications with rich functionality than ever.However,more functionality poses larger memory and storage space demands.Without proper management,user satisfaction may encounter severe issues: either because of frustrating application response time or limited memory capacity.Swapping method is an alternative way to extend memory capacity,thus enables system keeping more processes background.Mobile systems such as smartphones and tablets are re-adopting swapping to extend memory capacity without adding more DRAM.This resurgence of swapping in mobile systems has inspired both traditional ‘‘off-the-rack’’ schemes and new approaches based on compression and new hardware.Their vastly different designs,however,make them difficult for system designers to measure,compare and revise.To make this easy and improve user experience of mobile systems,we have done work as following.First,we propose an evaluation framework,SwapBench,to appraise swap schemes and focus on two important but overlooked metrics: application launch and switch.SwapBench is capable of detecting and configuring swapping,controlling and recording application execution procedure.And cross-validation with microbenchmarks shows that SwapBench is accurate.Then,we present the first comprehensive evaluation from three dimensions: system architecture,application launch time and application switch delays,to understand and summarize the impacts of swapping in mobile systems.Finally,based on the findings from SwapBench,we give our conclusion and suggestions of different approaches to swapping in mobile systems.Second,based on the findings with SwapBench,we further conduct a comprehensive measure-based study to reveal the reason of poor performance brought by the traditional flash-based swapping scheme in mobile device.Furthermore,we present a predictive process-level swapping method that predicts the most rarely used(MRU)application list and dynamically swap processes from the MRU list ahead-of-time to preserve free memory for future use.We conduct a comprehensive measurement to evaluate Smart Swap.Trace-based evaluations show the prediction accuracy of MRU application is above 90% and up to 100%.Utilizing SwapBench,SmartSwap shows up to 30% application launch performance increase compared to worst case brought by flash swap.And system could retain more than 50% processes background.Besides,each single ahead-of-time swap-out operation needs energy cost of no more than 0.03% battery life.