FindOldestPenny

PD_ParallelDecomposition PD_ParallelAlgorithms PD_ParallelPerformance TCPP_Algorithms TCPP_Programming CS0 CS1 CS2 DSA Systems touch visual

This activity was originally proposed by Mary Smith and Srishti Srivastava, with consultation from Brett A Becker.

Paper (Srivastava2019) and slides available on-line.

Similar Activities

FindSmallestCard, FindYoungestStudent

Details

See paper for details. In summary, the goal is to find the oldest penny in a pile of pennies. This is first done sequentially, with one student trying to determine the oldest penny in a group of 50 pennies. Next, two students are employed, with each student getting a bag of 25 pennies. Lastly, four students get two pennies each, while a fifth student gets 42 pennies. This last scenario indicates the importance of load-balancing, and Amdahl’s Law.

This activity is very similar to FindYoungestStudentinClass proposed by Chitra P (Chitra2019)

CS2013 Knowledge Unit Coverage

PD/Parallel Decomposition Core Tier 1

2. Identify opportunities to partition a serial program into independent parallel modules. [Familiarity]

PD/Parallel Algorithms Core Tier 2

4. Identify independent tasks in a program that may be parallelized. [Usage]

PD/Parallel Performance

1. Detect and correct a load imbalance. [Usage]

2. Calculate the implications of Amdahl’s law for a particular parallel algorithm [Usage]

TCPP Topics Coverage

Programming Topics

Algorithms Topics

Accessibility

This activity may be difficult for students who are blind. For those students, the similar activity by Chitra P (FindYoungestStudent) may be a better choice.

Assessment

Paper (Srivastava2019) performs assessment on this activity and another activity (MoreProcessorsNotAlwaysBetter). For this particular activity, assessment was performed on 98 students (73 male, 25 female) over four institutions. The vast majority (81.6%) were aged 18-22. In addition, 38 students were Freshmen, 18 were sophomores, 25 were juniors, and 17 were seniors. Students were given the ASPECT survey that measures student engagement related to three constructs (value of activity, instructor contribution, and personal effort). Statistial significance was measured using a combination of one-way and two-way ANOVA tests. While no statistical difference was discovered for any of the three constructs over gender or year in college, 23-27 year-olds had a statistically greater value of the activity and sense of instructor contribution than 18-22 year olds. The authors suggest that that younger students may be more used to active learning strategies from K-12, while it may be more novel for older college students.

Citations