Show HN: CuSBF – Faster GPU Bloom Filter for Sequence Data

Category: library

Tags: gpu, bloom-filter, bioinformatics, cuda, k-mer, sequence-analysis

Score: 7.5/10 (Innovation: 8, Technical: 9, Documentation: 7, Utility: 6)

cuSBF is a high-performance GPU implementation of the Super Bloom filter, leveraging minimizer-based shard selection and the findere scheme to dramatically accelerate batch k-mer insertion and query for sequence data. Its innovative combination of GPU parallelism with bioinformatics-specific optimizations achieves 60-230x speedups over CPU implementations, making it highly interesting for computational genomics and large-scale sequence analysis.

Target audience: bioinformaticians, computational biologists, data engineers

Repository: https://github.com/tdortman/cuSBF · Cuda · BSL-1.0

View on Hacker News