Syllabus CSCI 2400 Computer Systems

CSCI 2400: Computer Systems

Fall 2017

Department of Computer Science, University of Colorado at Boulder

Course Summary

This course introduces students to Computer Systems. The course will cover key aspects of how a software program executes on a modern computer, namely modern CPU hardware. In particular, the following major concepts will typically be studied:

Computer Architecture
Binary Representations, e.g. Two's Complement
Binary Arithmetic and Overflow
Assembly Language Programming (some 32-bit and mostly 64-bit x86) with Integers

Data Movement between CPU Registers and Memory
Integer Arithmetic
Control Operations
Arrays
Data Structures

IEEE Floating Point Representation
Stack and/or Frame Pointers and their relation to Function Calls
Buffer Overflow
Instruction-level Parallelism
Pipelining
Caching
Performance Optimization of Software Programs
Linking
Exceptional Control Flow, e.g. Signal Handling and Shells
Virtual Memory
Dynamic Memory Allocation on the Heap, e.g. malloc()

In addition, the class will gain familiarity with important software tools such as debuggers, compilers, editors, and virtual machines.

General Information

See the Moodle class Web page at http://moodle.cs.colorado.edu

Schedule & Location: MWF 2-2:50 pm in Chem 140
Course number: CSCI 2400. See also the registrar's Web site.
Prerequisites: CSCI 2270.
Instructors: Professor Rick Han
Offices: ECCR 1B05F (CS Systems Lab in Engineering Center)
Office Hours: Wed 10-12 noon (knock on door of systems lab to be let in, or call office phone 303-492-0914). Additional appointments as needed.

Email: rhan@cs.colorado.edu
Phone: 303-492-0914
Textbook:

Computer Systems: A Programmer's Perspective, Randal Bryant and David O'Hallaron, 3rd edition, Pearson, 2015.

It is important that you use the 3rd edition (focuses on 64-bit topics), not the 2nd edition.

Other useful references:

C: A Reference Manual (5th Edition) (Paperback) by Samuel P. Harbison and Guy L. Steele, Prentice Hall, 2002.
The C Programming Language, Brian Kernighan and Dennis Ritchie, 2nd edition, Prentice Hall, 1988.

TAs: Sean Donohoe, Sepideh Goodarzy, John Klingner, Zhiyuan Liu, Avinash Ratnavel, Tao Ruan, Yichen Wang
TA Office/Hours: Weekly recitation sections, and more as needed. See moodle for announcements.
TA Email: {Sean.Donohoe, Sepideh.Goodarzy, John.Klingner, Zhiyuan.Liu, Avinash.Ratnavel, Tao.Ruan,Yichen.Wang}@colorado.edu

class Web site:
See the Moodle class Web page at http://moodle.cs.colorado.edu. Assignments, lecture slides, and announcements can be found there. The moodle has a variety of useful features, including a forum for posting questions. Each student should establish an account on the moodle and then subscribe to our class on the moodle using the special enrollment key given out in class.

CSEL Accounts:
The Computer Science department maintains the Computer Science Educational Labs (CSEL) in the first floor of the ECCS Wing of the Engineering Center. The CSEL consists of a cluster of primarily Linux machines. It is a convenient facility for doing your lab assignments. If you are officially enrolled in this course, then you should be automatically signed up for an account in the CSEL, and will also automatically be given Buff One card key access to the CSEL.

For help with the CSEL or moodle, email help@cs.colorado.edu. CSEL updates are at https://twitter.com/CUBoulderCSEL.

Class VM:
Instructions for downloading and installing the class VM will be posted on the moodle and gone over in the first recitation. You can download the VM from https://foundation.cs.colorado.edu/vm/.

Course Assistants:
Justin Cai, Cassidy Haas, and Matthew Kaiser will our CAs offering assistance with 2400 in the CSEL at hours posted on the moodle.

Grading

The lab assignments constitute 40% of your grade.
Two to three midterms exams and the final exam are together worth 50% of your grade.
The quizzes are worth 5%.
The recitation questions make up the remaining 5%.

Grading for Lab Assignments:

Your primary assignments will be your "Lab Assignments," given every 2-3 weeks, each of which will be followed by a grading meeting to review your solution with the TA. The grade meetings are scheduled on the Moodle site typically just before the assignment is due, and will begin after the due-date of each Lab.

The grades for each lab will be based 40% upon the Task Success (i.e. "does it work") and 60% upon your explanation of your code/assignment and answering questions about the lab and its concepts. Historically speaking, students that have completed the assignment themselves usually have little problem passing the Q&A portion of the grade.

Students may work in teams of up to two for the labs only, but each student will still be responsible for scheduling their own grading meeting with the TA for each lab. You may help others only to the extent of answering typical questions that arise during compiling, debugging, and executing your lab assignments.

All assignments are due by the deadline stated. Extensions will not be granted except at the instructor's discretion in documented cases of extreme hardship, emergency, etc., unless otherwise noted.

On the task success 40% portion of the grade, we strongly encourage you to submit even partially working code/assignment by the deadline to obtain partial credit on the 40% for task success.
On the Q&A 60% portion of the grade, you must attend your grading meeting to qualify for Q&A grading points. If you miss your meeting with the TA (without notifying your TA ahead of time with a suitable reason), this may result in a zero grade for 60% Q&A portion of the assignment. The TA is under no obligation to reschedule your appointment if you miss your meeting, so write down your meeting times, and don't forget them! Even if you are unable to submit fully working code/assignment by the deadline, we strongly encourage you to keep working at a full solution for the assignment, which should benefit your understanding and ability to answer questions during the Q&A meeting with the TA.

All labs must be written in C and compiled for execution on the class VM, unless otherwise noted.

More information on the grading policy may be posted as needed on the moodle as the semester progresses.

Default Grading Policy:

All work should be your own on all assignments unless otherwise indicated.

Plagiarism policy.

Additional Policies:

Disability Policy

Religious Observances Policy

Discrimination and Sexual Harassment Policy

Classroom Behavior Policy