CIS SIGSYS Vivek K. Pallipuram
Title: Exploring Multiple Levels of Performance Modeling for Heterogeneous Systems
Name: Dr. Vivek K. Pallipuram
Abstract
One of the major challenges faced by the HPC community today is user-friendly and accurate heterogeneous performance modeling. Heterogeneous systems such as the General Purpose Graphical Processing Unit (GPGPU) clusters are now ubiquitous due to their massive computational capabilities; however these resources may be wasted due to inefficient application mapping and tuning. Although performance prediction models exist to fine-tune applications, they are seldom easy-to-use and do not address multiple levels of design space abstraction. Our research aims to bridge the gap between reliable performance model selection and user-friendly analysis. We propose a straightforward and accurate multi-level performance modeling suite for heterogeneous systems that addresses multiple levels of design space abstraction. The multi-level performance modeling suite primarily targets synchronous iterative algorithms (SIAs) using oursynchronous iterative GPGPU execution (SIGE) model and addresses two levels of design space abstraction: 1) low-level where partial details of the implementation are present along with system specifications and 2) high-level where implementation details are minimum and only high-level system specifications are known. The low-level abstraction of the modeling suite employs statistical techniques for runtime prediction, whereas the high-level abstraction utilizes existing analytical and quantitative modeling tools to predict the application runtime. Our verification efforts for the low-level and high-level abstraction models yield high runtime prediction accuracy with less than 10% error rate for several tested GPGPU cluster configurations and case studies. The end goal of our research is to offer the scientific community, a reliable and user-friendly performance prediction framework that allows them to optimally select a performance prediction strategy for the given design goals and system architecture characteristics.
Brief Biography
Vivek earned his Bachelor of Technology degree in Instrumentation and Control Engineering from National Institute of Technology Trichy, India in 2008, Master of Science in Computer Engineering in 2010, and Doctor of Philosophy (Ph.D.) in Computer Engineering in 2013; both from Clemson University, SC USA. His research interests are in the broad field of high-performance computing (HPC). These interests include, but are not limited to developing performance prediction models for heterogeneous systems, evaluating emerging HPC architectures and programming paradigms, designing algorithms for parallel computing, and optimizing codes for HPC resources. Currently, Vivek is working with Dr. Michela Taufer as a post-doctoral researcher and exploring significant aspects of computing such as HPC on clouds and dynamically heterogeneous systems.
Tuesday, March 18, 2014 at 3:30am to 4:45am
Smith Hall, 102A
Smith Hall, University of Delaware, Newark, DE 19716, USA
- Event Type
- Calendar
- Departments
- Website
- Group
- ENGR - Computer & Information Sciences
- Contact Email
- Contact Name
-
Steve Siegel
- Subscribe
Recent Activity
No recent activity