Optimizing Bloom Filters for Modern GPU Architectures

TL;DR

This paper presents an optimized GPU-based Bloom filter design that significantly improves throughput and accuracy, outperforming existing methods by over 11 times while maintaining high precision and leveraging GPU architecture features.

Contribution

The authors develop a novel GPU Bloom filter implementation that balances speed and accuracy, exploring design space dimensions like vectorization and thread cooperation.

Findings

Achieves over 11x speedup in lookup and construction compared to state-of-the-art.

Maintains high accuracy while delivering high throughput close to GPU speed limits.

Optimizations are most effective when the filter fits within GPU cache.

Abstract

Bloom filters are a fundamental data structure for approximate membership queries, with applications ranging from data analytics to databases and genomics. Several variants have been proposed to accommodate parallel architectures. GPUs, with massive thread-level parallelism and high-bandwidth memory, are a natural fit for accelerating these Bloom filter variants potentially to billions of operations per second. Although CPU-optimized implementations have been well studied, GPU designs remain underexplored. We close this gap by exploring the design space on GPUs along three dimensions: vectorization, thread cooperation, and compute latency. Our evaluation shows that the combination of these optimization points strongly affects throughput, with the largest gains achieved when the filter fits within the GPU's cache domain. We examine how the hardware responds to different parameter configurations and relate these observations to measured performance trends. Crucially, our optimized design overcomes the conventional trade-off between speed and precision, delivering the throughput typically restricted to high-error variants while maintaining the superior accuracy of high-precision configurations. At iso error rate, the proposed method outperforms the state-of-the-art by $11.35\times$ ($15.4\times$) for bulk filter lookup (construction), respectively, achieving above $92\%$ of the practical speed-of-light across a wide range of configurations on a B200 GPU. We propose a modular CUDA/C++ implementation, which will be openly available soon.