Accelerating string matching for bio-computing applications on multi-core CPUs

Abstract

Huge amount of data in the form of strings are being handled in bio-computing applications and searching algorithms are quite frequently used in them. Many methods utilizing on both software and hardware are being proposed to accelerate processing of such data. The typical hardware-based acceleration techniques either require special hardware such as generalpurpose graphics processing units (GPGPUs) or need building a new hardware such as an FPGA based design. On the other hard, software-based acceleration techniques are easier since they only require some changes in the software code or the software architecture. Typical software-based techniques make use of computers connected over a network, also known as a network grid to accelerate the processing. In this paper, we test the hypothesis that multi-core architectures should provide better performance in this kind of computation, but still it would depend on the algorithm selected as well as the programming model being utilized. We present the acceleration of a stringsearching algorithm on a multi-core CPU via a POSIX thread based implementation. Our implementation on an 8-core processor (that supports 16-threads) resulted in 9x throughput improvement compared to a single thread implementation.

Combinatorial optimization of DNA sequence analysis on heterogeneous systems

“…For performance comparison, we use the absolute difference and the percent difference, absolute_d iff erence D jt EM t SA j percent_d iff erence D 100 absolute_d iff erence=t EM (7) where t EM indicates the best execution time determined using EM and t SA indicates the execution time of our algorithm with a system configuration suggested by the SA approach. Figure 8 depicts the execution time and the standard deviation of our DNA sequence analysis implementation using the system configuration suggested by the SA for various types of DNA sequences.…”

“…Herath et al . presented an implementation of the Aho–Corasick (AC) algorithm based on pattern partitioning. A prefix‐based input partitioning approach is presented by Drews et al .…”

Analyzing Large-Scale DNA Sequences on Multi-core Architectures

“…Herath et al presented in [16] an implementation of the Aho-Corasick string matching algorithm using POSIX threads, which is based on the pattern partitioning approach. A replication of the Herath's study with the intention to improve the software implementation of the Aho-Corasick algorithm was conducted by Arudchutha et al [6].…”

“…The hardware based approaches (such as [7], [27]) are faster, but less flexible and more expensive, whereas software based acceleration techniques are flexible in terms of updating or adding new patterns [30]. Recently different software based DNA analysis techniques designed for multi-core systems have been proposed [6], [11], [14], [16], [19], [23].…”

String matching with multicore CPUs: Performing better with the Aho-Corasick algorithm

“…Many researches have been done utilizing both hardware and software to accelerate string matching in several areas: hardware supported approaches use FPGA [7], [12], GPU [3], [8], [9] and Cell/B.E. processor [5] and software based approaches use multiple processors [2], [4]. Among them, the software based acceleration techniques need only some modification in the software code or the architecture.…”