Capacity Planning for Software Applications in Natural Disaster Scenarios

Abstract

In recent years, there has been a notable increase in the development of social media andhumanitarian applications based on bots, designed to provide assistance during large-scalenatural disasters. These applications play a crucial role in managing the chaos and usefulto satisfy the urgent needs for rescue and relief that arise when catastrophes disruptdaily routines. However, they often encounter challenges during emergencies, such asdynamic and unpredictable variations in user workload, which can affect service qualityand application stability. To tackle these challenges, we propose a capacity planningmethodology to determine the optimal number of replicas and partitions for each componentof an application and distributes them across virtual machines in server clusters. Bybridging the gap between the algorithms executed in the applications and the performancecharacteristics of their implementations, this methodology enables applications to scaleefficiently. It helps maintain response times and average utilization within user-definedranges while providing fault tolerance to prevent component saturation. We validate theproposed methodology with tree bot-based applications devised to be use after a naturaldisaster strikes. Our experimental results show the effectiveness of the methodology, withestimation errors ranging from 1% to 15% for utilization and average response times. Furthermore,the methodology serves as an effective elasticity tool, allowing for the adjustmentof component replicas based on user’s requests.