Terascale Query Processing in the Browser: Rethinking GPU Acceleration

2026-07-20

AI Summary

This work addresses the challenge of efficiently executing compute-intensive recursive database queries directly within web browsers, a task poorly supported by existing systems. To this end, the authors propose WGLog, the first browser-native GPU-accelerated engine leveraging WebGPU compute shaders. Its key innovations include replacing conventional hash-join operations with atomic-free, ordered array joins and integrating WebGPU’s indirect dispatch mechanism to construct an asynchronous execution pipeline, thereby eliminating synchronization overhead and serialization bottlenecks. Experimental evaluation demonstrates that WGLog achieves 1.48× to 4.68× speedups over native GPU-based systems on representative workloads and delivers orders-of-magnitude performance improvements compared to CPU and WebAssembly implementations.

Abstract

Recursive query computation, central to graph algorithms and relational databases, demands GPU acceleration due to its inherent computational intensity. While substantial prior work addresses GPU implementations of recursive queries that require fixed-point evaluation, existing systems are restricted to native execution environments. We introduce WGLog, the first web-browser-native GPU engine for compute-bound recursive database queries. WGLog is built entirely on WebGPU compute shaders, a cross-platform API that enables GPU acceleration in web browsers. WGLog leverages two key technical innovations. First, we replace hash-table-based joins with atomic-free sorted-array joins, eliminating the serialization bottleneck that hash tables suffer on skewed graphs. Second, we develop an asynchronous execution pipeline using WebGPU's indirect dispatch capability, which eliminates GPU-host synchronizations that would otherwise dominate per-iteration overhead. On representative workloads, WGLog delivers a 1.48--4.68x speedup over native GPU systems and orders-of-magnitude improvement over CPU and WebAssembly implementations.

Problem

• recursive queries
• GPU acceleration
• web browser
• database query processing
• WebGPU

Innovation

• WebGPU
• recursive queries
• sorted-array joins
• asynchronous execution
• browser-native GPU