Links

Tools

Export citation

Search in Google Scholar

SLICEE: A Service oriented middleware for intensive scientific computation

This paper is available in a repository.
This paper is available in a repository.

Full text: Download

Question mark in circle
Preprint: policy unknown
Question mark in circle
Postprint: policy unknown
Question mark in circle
Published version: policy unknown

Abstract

With the growth of computational power and storage demand from bioinformatics applications, cluster computing has emerged as a good answer. Bioinformatics applications are often structured as workflows that are composed of a set of operations to perform on large data sets. These workflows are deployed as complex script that handles the sequence of program calls with their relevant inputs and try to take advantage of a computer cluster using a scheduler. Their performance rely on the user ability to analyze the potential parallelism in the workflow. SLICEE (Service Layer for Intensive Computation Execution Environment) proposes to abstract the calls to the cluster scheduler by handling command submission and data management. A workflow client is used to orchestrate the SLICEE services, that takes care of exploiting the data parallelism, and data routing between tasks to take advantage of the cluster architecture. A storage management layer takes care of optimizing data transfer with a reference passing mechanism. More-over SLICEE handles multi site command submission and data storage in a way that makes it transparent to the end-user. This tool enables an easy to maintain and to share implementation of bioinformatics workflows using intensive computation resources.