Mondays and Wednesdays 12:00 noon to 1:15 pm, Swearingen 2A31, starting January 13.
Stephen Fenner
Telephone: 803-777-2596 (only if I'm there; email is preferred)
Email: I use gmail, and my user name is fenner.sa
Web: http://www.cse.sc.edu/~fenner
Office: Swearingen 3A65, or via email
Office Hours: MW 1:30 pm - 2:30 pm, 4:10 pm - 5:00 pm.
CSCE 212, MATH 242, STAT 509
Required: M. Harchol-Balter, Performance Modeling and Design of Computer Systems: Queueing theory in action. Cambridge, 2013.
Lecture topics, exercises, and exams are all tied closely with the textbook.
The course description is, to quote the catalog:
System-level modeling and evaluation of computer systems: requirements elicitation and specification, architectural design, reliability and performance evaluation, Markov modeling, life-cycle cost analysis, project management.Complex computer systems comprise a variety of components (CPUs, disks, routers, and other devices), all interacting with each other, perhaps asynchronously. Modeling and analyzing the performance of these systems is crucial for good system design and optimization. This course introduces the mathematical tools used to model and predict system behavior---principally, queueing theory and analysis of stochastic processes---and applies them to a number of realistic scenarios.
The course maintains a "big picture" approach to computer systems, viewing them over their entire lifecycle.
The course is mathematical in nature, and students must apply mathematical concepts and results to specific situations as well as derive mathematical laws.
Quoting the departmental syllabus for the course, the outcomes are to:
We will touch on all these but will concentrate on the third point above.
The work for the course includes
The first midterm exam is on Wednesday, February 19; the second is on Monday, March 31.
Each midterm exam takes up one lecture period (1 hour, 15 minutes) in the classroom, and is open book, open notes, but no electronic devices, except for legitimate use by students with documented disabilities.
The final exam is 2.5 hours on Friday, May 2, from 4:00pm to 6:30pm in the classroom. The format and rules of the final exam are the same as for the midterm exams.
PLEASE NOTE:
Letter grades correspond to score percentages as follows:
| A | 90 or above |
| B+ | at least 86 and less than 90 |
| B | at least 80 and less than 86 |
| C+ | at least 76 and less than 80 |
| C | at least 70 and less than 76 |
| D+ | at least 66 and less than 70 |
| D | at least 60 and less than 66 |
| F | less than 60 |
You (the student) are expected to read all assigned material before the lecture begins. If for some reason you cannot attend class, you are responsible for any material covered during your absence.
No make-up exams will be given except under extreme circumstances with a valid excuse, in which case you must give me notice well before the exam if at all possible. Valid excuses include grave illness or death in the immediate family, or other dire emergency. They do not include car problems or planned, non-emergency events (weddings, trips, etc.).
I will maintain an email list for the class to which I will send out announcements from time to time, each with a subject line starting with "CSCE 317:". To receive these emails, you must send me an email (from an account where you want to receive these messages) with the subject "317 mail list" (the message body can be empty).
You are expected to know the Academic Code of Responsibility as it appears in the Carolina Community: Student Policy Manual.
You may not represent other people's work as your own. More specifically, you cannot submit specific answers or computer programs from other sources without proper attribution. If you copy or otherwise derive materials/answers from other people, books, the web, etc., you must cite your source(s) in a way that adheres to the usual standards for an academic paper. Deriving/copying without proper attribution is plagiarism, which I regard as a serious offense. (Exceptions: You do not need to explicitly cite any material you lift from the text or from my own lectures. You are also not required to explicitly cite any general background material you use for an answer but which does not specifically relate to the actual question being answered.)
Obtaining answers during an exam other than by the approved means (your own printed material, knowledge, and cleverness) is forbidden. Passing along questions or answers to an exam to anyone else before the exam is given is also forbidden.
Any violation of the rules above constitutes cheating, for which there is no excuse.
THE USUAL PENALTY FOR CHEATING IS FAILURE OF THE COURSE. The offense will also be reported to the appropriate University entities. The bare minimum penalty you may receive for an instance of cheating is double the points of the assignment off. That means that if an assignment/test is worth 10 points, your ultimate grade would be -20 for the assignment/test. Finally, as noted in the Student Policy Manual, the maximum penalty for cheating on an assignment is expulsion from the University. These penalties apply both to copier and copiee.
Course topics are given below.
This schedule is flexible and is subject to some revision as the class proceeds. There may not be time to cover all the topics listed. You will be tested only on those topics that could be covered in class, which means that the content of the exams may be adjusted. The topics to be covered in each midterm exam will be determined as the exam approaches.
At a minimum, I would like to get through Part IV of the text, perhaps omitting some material along the way, particularly in Chapters 5 and 9. The schedule below reflects this goal, and amounts to about one chapter per week. Time permitting, some topics in later parts of the book may also be covered.
| Topic(s) | Chapter | Week(s) |
| Introduction to queueing theory | 1 and 2 | 1 - 2 |
| Probability review | 3 | 3 |
| Generating random variables | 4 | 4 |
| Sample paths, convergence, and averages | 5 | 5 |
| Operational laws | 6 | 6 |
| Modification analysis | 7 | 7 |
| Discrete-time Markov chains | 8 | 8 - 9 |
| Ergoticity theory | 9 | 10 |
| Examples: Google, Aloha, and arder chains | 10 | 11 |
| Poisson process | 11 | 12 |
| Continuous-time Markov chains | 12 | 13 |
| M/M/1 and PASTA | 13 | 14 |
Last updated Monday January 13, 2014 at 10:32:29 EST.