Optimizing cloud storage costs using intelligent data tiering algorithms

Abstract

Cloud storage has become indispensable for enterprises, but escalating costs remain a significant challenge, particularly when files are allocated inefficiently across storage tiers. This study investigates the effectiveness of intelligent data tiering algorithms for optimizing cloud storage costs. Using a simulated dataset of 10 TB collected over six months, we designed and implemented tiering policies that dynamically reallocated files based on access frequency. Cost modeling was performed in Microsoft Excel 2021, and statistical analyses, including descriptive statistics, ANOVA, regression modeling, and t-tests, were conducted using IBM SPSS Statistics 28. The results demonstrate that intelligent tiering reduced costs by 28.6% compared to baseline static allocation, with ANOVA confirming significant cost differences between tiers (p = 0.003). Regression analysis revealed a negative relationship between access frequency and storage cost, while t-tests verified the superiority of intelligent tiering over conventional approaches (p = 0.001). Validation across expanded datasets confirmed the algorithm’s robustness and scalability. These findings suggest that intelligent data tiering provides a cost-effective, statistically reliable, and adaptable framework for managing enterprise cloud storage.