High performance computing methods for the integration and analysis of biomedical data using SAS

Justin R. Brown, Valentin Dinu

Research output: Contribution to journalArticlepeer-review

7 Scopus citations

Abstract

From microarrays and next generation sequencing to clinical records, the amount of biomedical data is growing at an exponential rate. Handling and analyzing these large amounts of data demands that computing power and methodologies keep pace. The goal of this paper is to illustrate how high performance computing methods in SAS can be easily implemented without the need of extensive computer programming knowledge or access to supercomputing clusters to help address the challenges posed by large biomedical datasets. We illustrate the utility of database connectivity, pipeline parallelism, multi-core parallel process and distributed processing across multiple machines. Simulation results are presented for parallel and distributed processing. Finally, a discussion of the costs and benefits of such methods compared to traditional HPC supercomputing clusters is given.

Original languageEnglish (US)
Pages (from-to)553-562
Number of pages10
JournalComputer Methods and Programs in Biomedicine
Volume112
Issue number3
DOIs
StatePublished - Dec 2013

Keywords

  • High performance computing
  • Parallel processing
  • SAS

ASJC Scopus subject areas

  • Software
  • Computer Science Applications
  • Health Informatics

Fingerprint

Dive into the research topics of 'High performance computing methods for the integration and analysis of biomedical data using SAS'. Together they form a unique fingerprint.

Cite this