Terabyte-Scale Analytics in the Blink of an Eye

2025-06-10

AI Summary

This work addresses CPU architectural bottlenecks in TB-scale analytical workloads by exploring the theoretical and practical limits of GPU cluster acceleration for distributed SQL query processing. We propose the first GPU-native distributed SQL execution framework tailored for OLAP workloads, innovatively integrating NCCL collective communication primitives with asynchronous RDMA-based data movement, and introducing a columnar memory layout coupled with query-level pipelined scheduling. We provide the first systematic scalability evaluation of GPU clusters under analytical workloads, establishing an upper bound of over 60× end-to-end performance improvement. On the full TPC-H 1 TB benchmark—comprising all 22 queries—we achieve sub-400 ms completion time (roughly two human eye blinks) and attain 60× higher throughput than an equivalently configured CPU cluster. This work delivers a foundational architectural paradigm and empirical benchmark for GPU-native distributed analytical databases.

Technology Category

• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems
• Search and Optimization: Distributed Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Querying, indexing, and retrieval in Web-related graphs
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

For the past two decades, the DB community has devoted substantial research to take advantage of cheap clusters of machines for distributed data analytics -- we believe that we are at the beginning of a paradigm shift. The scaling laws and popularity of AI models lead to the deployment of incredibly powerful GPU clusters in commercial data centers. Compared to CPU-only solutions, these clusters deliver impressive improvements in per-node compute, memory bandwidth, and inter-node interconnect performance. In this paper, we study the problem of scaling analytical SQL queries on distributed clusters of GPUs, with the stated goal of establishing an upper bound on the likely performance gains. To do so, we build a prototype designed to maximize performance by leveraging ML/HPC best practices, such as group communication primitives for cross-device data movements. This allows us to conduct thorough performance experimentation to point our community towards a massive performance opportunity of at least 60$ imes$. To make these gains more relatable, before you can blink twice, our system can run all 22 queries of TPC-H at a 1TB scale factor!

Problem

• Scaling analytical SQL queries on GPU clusters
• Establishing upper bounds on performance gains
• Leveraging ML/HPC practices for data movement

Innovation

• Leveraging GPU clusters for distributed SQL analytics
• Using ML/HPC best practices for performance maximization
• Implementing group communication primitives for data movement