Optimizing I/O for Data‑Intensive Analytics in Cloud‑Native Environments: Insights from Uber Presto

Content Overview

This article explores the widespread industry shift of migrating data‑intensive analytical applications to cloud‑native environments, emphasizing that the unique cost model of cloud storage demands a more nuanced understanding of performance optimization.

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

Key findings include that over 50% of data accesses are smaller than 10 KB and more than 90% are under 1 MB, indicating a highly fragmented access pattern that has different implications for cloud versus on‑premise platforms.

The paper presents a case‑study‑driven logical framework and strategies for I/O optimization tailored to cloud environments, aiming to help readers design cost‑effective I/O solutions for data‑intensive applications.

Core Chapters

• Adjust cognition and strategy based on varying cloud storage scenarios and their impact on application design and performance.

• Using Uber data as a case study, illustrate the additional costs that common I/O optimization techniques may incur during cloud migration.

• Offer a fresh perspective on system design in the cloud computing domain to assist stakeholders in addressing the rapid growth of data‑intensive workloads.

Reader Benefits

The whitepaper provides a case‑study‑based presentation of I/O optimization logic and ideas, serving as a starting point for further research and helping readers craft efficient I/O strategies specifically for cloud‑native, data‑intensive applications.