Give

Graduate Exam Abstract


Ryan Friese

Ph.D. Preliminary
February 27, 2014, 8:00 am - 10:00 am
ISTeC Conference Room CS 305
Resource Management for Heterogeneous Computing Systems: Utility Maximization, Energy-Aware Scheduling, and Multi-Objective Optimization

Abstract: As high performance heterogeneous computing
systems continually become faster, the operating
cost to run these systems has increased. A
significant portion of the operating costs can be
attributed to the amount of energy required for
these systems to operate. To reduce these costs it
is important for system administrators to operate
these systems in an energy efficient manner.
Additionally, it is important to be able to
measure the performance of a given system so that
the impacts of operating at different levels of
energy efficiency can be analyzed. The goal of
this research is to examine how energy and system
performance interact with each other for a variety
of environments. One part of the study considers a
computing system and its corresponding workload
based on the expectations for future environments
of Department of Energy and Department of Defense
interest. New heuristics are presented that
maximize a performance metric created using
utility functions. A framework has been
established to analyze the trade-offs between
performance (utility earned) and energy
consumption. Furthermore, the previous heuristics
have been adapted, as well as new heuristics and
energy filtering techniques have been designed for
a computing system that has the goal of maximizing
the total utility earned while being subject to an
energy constraint. Currently, stochastic
environment models where utility needs to be
maximized under a probabilistic energy constraint,
given uncertainties in the execution time and
energy consumption of tasks in the system is being
examined. In addition to using utility earned as
a measure of system performance, system makespan
has also been studied. The trade-offs between
minimizing makespan and minimizing energy for
various environments has been analyzed. Finally, a
framework is being developed that will enable the
investigation of the effects of P-states and
memory interference on energy consumption and
system performance.


Adviser: H.J. Siegel
Co-Adviser: Anthony Maciejewski
Non-ECE Member: Patrick Burns, Mechanical Engineering
Member 3: Sudeep Pasricha, Electrical and Computer Engineering
Addional Members: Greg Koenig, Oak Ridge National Lab

Publications:
Ryan Friese, Tyler Brinks, Curt Oliver, Howard Jay Siegel, and Anthony A. Maciejewski, “Analyzing the Trade-offs Between Minimizing Makespan and Minimizing Energy Consumption in a Heterogeneous Resource Allocation Problem,” The Second International Conference on Advanced Communications and Computation (INFOCOMP 2012), cosponsors: IARIA et al., pp. 81-89, Venice, Italy, Oct. 2012. Received one of seven “best paper” awards given.

Ryan Friese, Bhavesh Khemka, Anthony A. Maciejewski, Howard Jay Siegel, Gregory A. Koenig, Sarah Powers, Marcia Hilton, Jendra Rambharos, Gene Okonski, and Stephen W. Poole, “An Analysis Framework for Investigating the Trade-offs Between System Performance and Energy Consumption in a Heterogeneous Computing Environment,” 22nd Heterogeneity in Computing Workshop (HCW 2013), cosponsors: IEEE Computer Society and U.S. Office of Naval Research, in the proceedings of 2013 International Parallel and Distributed Processing Symposium Workshops and PhD Forum (IPDPSW), pp. 19-30, Boston, MA, May 2013.

Ryan Friese, Tyler Brinks, Curt Oliver, Anthony A. Maciejewski, Howard Jay Siegel, and Sudeep Pasricha, “A Machine-by-Machine Analysis of a Bi-Objective Resource Allocation Problem,” The 2013 International Conference on Parallel and Distributed Processing Techniques and Applications (PDPTA 2013), Vol. I, sponsor: World Academy of Science and Computer Science Research, Education, and Applications (CSREA), pp. 3-9, Las Vegas, NV, July 2013.

Kyle M. Tarplee, Ryan Friese, Anthony A. Maciejewski, and Howard Jay Siegel, “Efficient and Scalable Computation of the Energy and Makespan Pareto Front for Heterogeneous Computing Systems,” 6th Workshop on Computational Optimization (WCO ‘13), part of the 8th Symposium on Advances in Artificial Intelligence and Applications, in the proceedings of the Federated Conference on Computer Science and Information Systems (FedCSIS 2013), cosponsors: Polish Ministry of Science and Higher Education and Intel, pp. 401-408, Krakow, Poland, Sep. 2013. Received “The 2013 Zdzislaw Pawlak Best Paper Award, by the Award Committee of the 8th Symposium on Advances in Artificial Intelligence and Applications, for the paper ‘Efficient and Scalable Computation of the Energy and Makespan Pareto Front for Heterogeneous Computing Systems’."

Paul Maxwell, Anthony A. Maciejewski, Howard Jay Siegel, Jerry Potter, Gregory Pfister, Jay Smith, and Ryan Friese, “Robust Static Planning Tool for Military Village Search Missions: Model and Heuristics,” Journal of Defense Modeling and Simulation, Vol. 10, No. 1, pp. 31-47, Jan. 2013.

Paul Maxwell, Ryan Friese, Anthony A. Maciejewski, Howard Jay Siegel, Jerry Potter, and James Smith, “A Demonstration of a Simulation Tool for Planning Robust Military Village Searches,” Huntsville Simulation Conference (HSC '10), sponsor: The Society for Modeling & Simulation International, Huntsville, AL, Oct. 2010.

Ryan Friese, Paul Maxwell, Anthony A. Maciejewski, and Howard Jay Siegel, “A Graphical User Interface for Simulating Robust Military Village Searches,” International Conference on Modeling, Simulation and Visualization Methods (MSV ‘11), sponsor: World Academy of Science and Computer Science Research, Education, and Applications (CSREA), pp. 75-81, Las Vegas, NV, July 2011.


Program of Study:
CS 540
MATH 560
GRAD 544
ECE 555
CS645
ECE 795
ECE 799
N/A