An experimental and comparative benchmark study examining resource utilization in managed Hadoop context

Transitioning cloud-based Hadoop frameworks from IaaS to PaaS, which are commercially conceptualized as pay-as-you-go or pay-per-use, often reduces the associated system costs. However, managed systems obscure inner performance dynamics of platform and present a black-box behavior end-users. The aim this study was investigate resource utilization current platforms. Thus, we explored three prominent Hadoop-on-PaaS proposals they come out-of-the-box conducted Hadoop-specific workloads using HiBench Benchmark Suite. During benchmark executions, data worker nodes were collected analyzed. results indicated that same property specifications among cloud services neither do guarantee similar outputs, nor produce consistent based on different within themselves. We anticipate systems’ architectures pre-configurations play crucial role in outcomes.