Comparative Analysis of CPU vs. GPU Performance in Bioinformatics Data Processing

EasyChair Preprint 13778

Abstract

In the rapidly evolving field of bioinformatics, the need for efficient data processing has never been more critical. This study presents a comparative analysis of CPU (Central Processing Unit) and GPU (Graphics Processing Unit) performance in the context of bioinformatics data processing. Bioinformatics tasks, often characterized by their computational intensity and large data sets, provide a fertile ground for exploring the advantages and limitations of different hardware architectures.

The analysis focuses on key bioinformatics applications, including sequence alignment, genomic data analysis, and molecular dynamics simulations. We benchmark these applications on both CPU and GPU platforms, evaluating performance metrics such as processing time, energy consumption, and scalability. Our findings reveal that GPUs, with their parallel processing capabilities, significantly outperform CPUs in tasks that can be highly parallelized, such as sequence alignment and molecular dynamics simulations. Conversely, CPUs exhibit superior performance in tasks requiring complex control logic and lower levels of parallelism.

Furthermore, the study delves into the cost-effectiveness of deploying GPUs over CPUs in bioinformatics research and the practical considerations of integrating GPU-accelerated computing into existing bioinformatics workflows. The results underscore the potential of GPUs to revolutionize bioinformatics data processing, offering a path toward more efficient and scalable solutions. However, the study also highlights the importance of task-specific hardware optimization and the need for continued research into hybrid computing approaches that leverage the strengths of both CPUs and GPUs.

Keyphrases: Bio-informatics, CPU (Central Processing Unit), GPU (Graphics Processing Unit)

@booklet{EasyChair:13778,
  author    = {Abill Robert},
  title     = {Comparative Analysis of CPU vs. GPU Performance in Bioinformatics Data Processing},
  howpublished = {EasyChair Preprint 13778},
  year      = {EasyChair, 2024}}