Migrating Data‑Intensive Analytics to Cloud‑Native Environments: Cost‑Aware I/O Optimization Insights from Uber Presto

This article explores the widespread industry shift of migrating data‑intensive analytics applications from on‑premises to cloud‑native environments, emphasizing that the unique cost model of cloud storage demands a finer‑grained understanding of performance optimization.

Through empirical analysis of Uber's Presto production workload, the authors discover that traditional I/O optimizations often overlook the financial cost of storage API calls, which can lead to unexpectedly high expenses in cloud settings.

Observations show that Presto's data access patterns are highly fragmented: over 50% of accesses are smaller than 10 KB and more than 90% are under 1 MB, a characteristic that carries different implications for cloud versus traditional data platforms.

The paper presents a case‑study‑driven approach to I/O optimization, offering logical frameworks and strategies tailored for cloud environments, aiming to help readers design efficient I/O solutions that improve cost‑performance ratios for data‑intensive applications.

Overall, the whitepaper provides a fresh perspective on system design in the cloud computing domain, guiding stakeholders to address the rapid growth of data‑heavy workloads with cost‑aware optimization techniques.