[Colloquium] Reminder: Mishra/Dissertation Defense/Apr 25, 2017

[Margaret Jaffey via Colloquium]

This is a reminder about Nikita's defense tomorrow.

       Department of Computer Science/The University of Chicago

                     *** Dissertation Defense ***


Candidate:  Nikita Mishra

Date:  Tuesday, April 25, 2017

Time:  3:00 PM

Place:  Ryerson 255

Title: Statistical Methods for Performance Estimation for Improving
Scheduling and Energy Minimization

Abstract:
This thesis is about using statistical methods for performance and
power estimation which would allow us to develop better scheduling
algorithms and also more energy efficient systems. In many
deployments, computer systems are underutilized – meaning that
applications have perfor- mance requirements that demand less than
full system capacity. Ideally, we would take advantage of this
under-utilization by allocating system resources so that the
performance requirements are met and energy is minimized. This
optimization problem is complicated by the fact that the per- formance
and power consumption of various system configurations are often
application – or even input – dependent. Thus, practically, minimizing
energy for a performance constraint requires fast, accurate
estimations of application-dependent performance and power tradeoffs.
We propose a set of algorithms for different scenarios to tackle this
problem. First, we propose LEO, a probabilistic graphical model-based
learning system that provides accurate online estimates of an
application’s power and performance as a function of system
configuration. This work mostly focused on the performance estimation
for single applications. As the second part of our work, we look into
the estimation for application’s performance when they are
co-scheduled with other applications. As, the third part of our work,
we design a system called CALOREE which allows the learnt models to be
combined witha controller so that the system is robust to dynamic
situations with changing resource requirement. Two central challenges
arise when allocating system resources to meet these conflicting
goals: (1) complex- ity—modern hardware exposes diverse resources with
com- plicated interactions—and (2) dynamics—performance must be
maintained despite unpredictable changes in operating environment or
input. Machine learning accurately predicts the performance of
complex, interacting resources, but does not address system dynamics;
control theory adjusts resource usage dynamically, but struggles with
complex resource interaction. We therefore propose CALOREE, a
combination of learning and control that automatically adjusts
resource usage to meet performance requirements with minimal en- ergy
in complex, dynamic environments.

Nikita's advisor is Prof. John Lafferty

Login to the Computer Science Department website for details,
including a draft copy of the dissertation:

 https://www.cs.uchicago.edu/phd/phd_announcements#nmishra

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Margaret P. Jaffey            margaret at cs.uchicago.edu
Department of Computer Science
Student Support Rep (Ry 156)               (773) 702-6011
The University of Chicago      http://www.cs.uchicago.edu
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=