Year

2018

Season

Fall

Paper Type

Master's Thesis

College

College of Computing, Engineering & Construction

Degree Name

Master of Science in Computer and Information Sciences (MS)

Department

Computing

NACO controlled Corporate Body

University of North Florida. School of Computing

First Advisor

Dr. Swapnoneel Roy

Second Advisor

Dr. Asai Asaithambi

Rights Statement

http://rightsstatements.org/vocab/InC/1.0/

Third Advisor

Dr. Roger E. Eggen

Department Chair

Dr. Sherif Elfayoumy

College Dean

Dr. William F. Klostermeyer

Abstract

Energy consumption by various modern symmetric key encryption protocols (DES,

3-DES, AES and, Blowfish) is studied from an algorithmic perspective. The work

is directed towards redesigning or modifying the underlying algorithms for these

protocols to make them consume less energy than they currently do. This research

takes the approach of reducing energy consumption by parallelizing the

consecutive memory accesses of symmetric key encryption algorithms. To achieve

parallelization, an existing energy complexity model is applied to symmetric key

encryption algorithms. Inspired by the popular DDR3 architecture, the model assumes

that main memory is divided into multiple banks, each of which can store

multiple blocks. Each block in a bank can only be accessed from a cache of its

own, that can hold exactly one block. However all the caches from different banks

can be accessed simultaneously. In this research, experiments are conducted to

measure the difference in energy consumption by varying the level of parallelization,

i.e. variations of, number of banks that can be accessed in parallel. The

experimental results show that the higher the level of parallelism, smaller is the

energy consumption.

Download

Share

COinS