Data Caching for Enterprise-Grade Petabyte-Scale OLAP

TL;DR

This paper presents the Alluxio local cache, an architectural optimization for petabyte-scale OLAP systems that reduces network I/O and improves data transfer efficiency in large-scale enterprise environments.

Contribution

Introduction of the Alluxio local cache tailored for petabyte-scale OLAP, integrated with systems like Presto and HDFS, and validated through three years of deployment at Uber and Meta.

Findings

Significant reduction in network I/O and API call pressure.

Improved data transfer efficiency in large-scale workloads.

Effective handling of skewed and fragmented data access patterns.

Abstract

With the exponential growth of data and evolving use cases, petabyte-scale OLAP data platforms are increasingly adopting a model that decouples compute from storage. This shift, evident in organizations like Uber and Meta, introduces operational challenges including massive, read-heavy I/O traffic with potential throttling, as well as skewed and fragmented data access patterns. Addressing these challenges, this paper introduces the Alluxio local (edge) cache, a highly effective architectural optimization tailored for such environments. This embeddable cache, optimized for petabyte-scale data analytics, leverages local SSD resources to alleviate network I/O and API call pressures, significantly improving data transfer efficiency. Integrated with OLAP systems like Presto and storage services like HDFS, the Alluxio local cache has demonstrated its effectiveness in handling large-scale, enterprise-grade workloads over three years of deployment at Uber and Meta. We share insights and operational experiences in implementing these optimizations, providing valuable perspectives on managing modern, massive-scale OLAP workloads.