A Rabin-Karp Implementation for Handling Multiple Pattern-Matching on the GPU

Abstract

The volume of digital information is growing at an extremely fast pace which, in turn, exacerbates the need of efficient mechanisms to find the presence of a pattern in an input text or a set of input strings. Combining the processing power of Graphics Processing Unit (GPU) with matching algorithms seems a natural alternative to speedup the string-matching process. This work proposes a Parallel Rabin-Karp implementation (PRK) that encompasses a fast-parallel prefix-sums algorithm to maximize parallelization and accelerate the matching verification. Given an input text T of length n and p patterns of length m, the proposed implementation finds all occurrences of p in T in O(m + q + n τ + nm q) time, where q is a sufficiently large prime number and τ is the available number of threads. Sequential and parallel versions of the PRK have been implemented. Experiments have been executed on p ≥ 1 patterns of length m comprising of m = 10, 20, 30 characters which are compared against a text string of length n = 2 27. The results show that the parallel implementation of the PRK algorithm on NVIDIA V100 GPU provides speedup surpassing 372 times when compared to the sequential implementation and speedup of 12.59 times against an OpenMP implementation running on a multi-core server with 128 threads. Compared to another prominent GPU implementation, the PRK implementation attained speedup surpassing 37 times.

Comparison of parallel central processing unit‐ and graphics processing unit‐based implementations of greedy string tiling algorithm for source code plagiarism detection

“…Efficient implementations of the Karp–Rabin string matching algorithm on the GPU can be found in recent studies 34,35 . The authors of Reference 34 suggest a pattern grouping method that reduces the number of comparisons between the hash values with efficient allocation to the GPU threads.…”

“…Efficient implementations of the Karp-Rabin string matching algorithm on the GPU can be found in recent studies. 34,35 The authors of Reference 34 suggest a pattern grouping method that reduces the number of comparisons between the hash values with efficient allocation to the GPU threads. A parallel GPU-based Rabin-Karp implementation 35 utilizes a fast-parallel prefix-sums algorithm to parallelize the calculation of hashes and accelerate match verification.…”

“…34,35 The authors of Reference 34 suggest a pattern grouping method that reduces the number of comparisons between the hash values with efficient allocation to the GPU threads. A parallel GPU-based Rabin-Karp implementation 35 utilizes a fast-parallel prefix-sums algorithm to parallelize the calculation of hashes and accelerate match verification. Although they are quite complex in nature and not directly intended for source code similarity detection, both approaches might help improve our GPU implementation in future research.…”

Multi-Pattern GPU Accelerated Collision-Less Rabin-Karp for NIDS

“…The key difference between (Nunes et al, 2020) implementation and our proposed implementation is that our proposed one uses our hashing function, parallel architecture, and finally, we skip the string validation phase.…”

Multiple exact string matching algorithm based on q-gram hash comparison for DNA sequences

“…Çoklu desen eşleştirmede arama süresini kısaltmak için literatürde cuckoo filtrelemeye dayalı çoklu desen eşleştirme algoritmalarının paralelleştirilmesi [7], Rabin Karp algoritmasının paralelleştirilmesi [8], iki aşamalı paralel çoklu desen eşleştirme algoritması [9], Boyer-Moore-Horspool algoritmasına dayalı paralel çoklu dizi eşleştirme algoritmasının geliştirilmesi [10] ve mükemmel hash fonksiyonunu kullanarak GPU üzerinde paralel çoklu dizi eşleştirme algoritmalarının geliştirilmesi [11] gibi paralel programlama destekli birçok çalışma gerçekleştirilmiştir.…”