Kernel Level Support for Workflow Patterns
Abstract
In the evolution of computing technology over the decades, file system capabilities have not grown in tandem to processing power. Today, scientific computing is highly data intensive and relies on workflows. Also, scientific computations that rely on inherent workflows do not have a kernel support for workflows, and are executed essentially in a batch processing model. A file system that includes native kernel functionalities to support workflow execution would address the issue of parallel processing. Such a file system would improve scientific computing performance. This thesis describes an approach we developed to add workflow functionality to the Linux kernel and native file system to help simplify the use of workflow management systems for scientific computing. The approach is compared to Kepler workflow management system.
Keywords
FUSE, Workflow Patterns, Kepler
Date of this Version
7-11-2011
Department
Computer and Information Technology
Department Head
Jeffrey L. Brewer
Month of Graduation
August
Year of Graduation
2011
Degree
Master of Science
Head of Graduate Program
Jeffrey L. Brewer
Committee Member 1
Dr. Thomas J. Hacker
Committee Member 2
Dr. John A. Springer
Committee Member 3
Dr. Brandeis H. Marshall