Asee peer logo

Work in Progress: Exploring Students' Misconceptions of Cache Memories

Download Paper |

Conference

2022 ASEE Annual Conference & Exposition

Location

Minneapolis, MN

Publication Date

August 23, 2022

Start Date

June 26, 2022

End Date

June 29, 2022

Conference Session

ERM: Problem Solving and Conceptual Understanding

Page Count

8

DOI

10.18260/1-2--40679

Permanent URL

https://peer.asee.org/40679

Download Count

289

Paper Authors

biography

Muhammad Suleman Mahmood University of Illinois at Urbana - Champaign

visit author page

Suleman Mahmood is a Ph.D. student in the Department of Computer Science at the University of Illinois, Urbana-Champaign. Suleman completed his MS in Computer Science from Lahore University of Management Sciences and BS in Electrical Engineering from University of Engineering and Technology. He is interested in exploring how students learn computer science concepts and developing tools to assist them in the learning process.

visit author page

biography

Hongxuan Chen University of Illinois at Urbana - Champaign

visit author page

Hongxuan Chen is a Ph.D. student in the Department of Computer Science at the University of Illinois, Urbana-Champaign, where he also completed his B.S. in Computer Science. He is broadly interested in how students learn computer science and broadening participation in computer science.

visit author page

biography

Morgan Fong

visit author page

Morgan M. Fong is a Ph.D. student in the Department of Computer Science at the University of Illinois, Urbana-Champaign and an NSF Graduate Research Fellow. Prior to starting her Ph.D. Morgan completed her B.A. in Computer Science at the University of California, Berkeley. Her current research focuses on developing methods for observing and analyzing cooperative learning in undergraduate computing courses.

visit author page

biography

Geoffrey Herman University of Illinois at Urbana - Champaign

visit author page

Dr. Geoffrey L. Herman is the Severns Teaching Associate Professor with the Department of Computer Science at the University of Illinois at Urbana-Champaign. He earned his Ph.D. in Electrical and Computer Engineering from the University of Illinois at Urbana-Champaign as a Mavis Future Faculty Fellow and conducted postdoctoral research with Ruth Streveler in the School of Engineering Education at Purdue University. His research interests include creating systems for sustainable improvement in engineering education, conceptual change and development in engineering students, and change in faculty beliefs about teaching and learning. He is an associate editor with the Journal of Engineering Education and a board member of the Computing Research Association Education committee.

visit author page

Download Paper |

Abstract

Memory is an important part of computer systems. It is a performance bottleneck for many applications. To get maximum performance from the memory system, programmers need to know how caches work and what data access patterns lead to better performance. Caches also play an important role in computer security. Caches are recognized as one of the most difficult topics in Computer Architecture courses. Our earlier investigations indicated that students struggle particularly in mapping data to cache and in keeping track of that state of set-associative caches and their replacement policies. In this paper we describe our ongoing exploration of the methods that students use to map data to cache and to track replacement polices for set associative caches. We also investigate the misconceptions that hinder the students in successfully accomplishing these tasks. We believe that this investigation will help us in designing representations and tools to help students develop a better understanding of caches.

We conducted a series of 10 think-aloud interviews in Spring 2020 semester. The students were asked to solve three questions about mapping data to cache and two questions about tracking least recently used replacement policy for set-associative caches. The interviews were transcribed and analyzed by a research team consisting of four members using constant comparative method to form our hypotheses. Constant comparative method is derived from grounded theory. It requires that the researchers constantly compare new data with emerging theories and prior data. This paper describes the motivation, research questions, data collection and analysis methodology that we have followed in our study along with our initial results.

Our preliminary findings indicate that students assume sequential access in mapping data to cache even when there is no access pattern mentioned in the questions. Our analysis showed misconceptions about how data is cached in the cache blocks. For tracking the state of the cache, students tend to ignore the replacement policy while placing data in a set-associative cache. A common observed pattern is that students go back to the beginning to recompute the state of the cache when students realize the need for the replacement policy rather than actively tracking least recently used state of the cache.

Some of our observations have limited support in the collected data. We are in the process of collecting more data to test our emerging hypotheses. The additional data will be analyzed through the same process.

Mahmood, M. S., & Chen, H., & Fong, M., & Herman, G. (2022, August), Work in Progress: Exploring Students' Misconceptions of Cache Memories Paper presented at 2022 ASEE Annual Conference & Exposition, Minneapolis, MN. 10.18260/1-2--40679

ASEE holds the copyright on this document. It may be read by the public free of charge. Authors may archive their work on personal websites or in institutional repositories with the following citation: © 2022 American Society for Engineering Education. Other scholars may excerpt or quote from these materials with the same citation. When excerpting or quoting from Conference Proceedings, authors should, in addition to noting the ASEE copyright, list all the original authors and their institutions and name the host city of the conference. - Last updated April 1, 2015