CS 441 - Computer Architecture

Meeting time: TR 9:45-11:15am
Room 206 Chapman Building

University of Alaska Fairbanks

UAF CS F441-F01
3.0 Credits, Fall 2010
Prerequisites: CS 321 (OS), EE 341

Instructor: Dr. O. Lawlor
lawlor@alaska.edu, 474-7678
Office: 201E Chapman
Hours: 11:30-1:00 TR, by appointment, or just drop by!

Recommended Textbook:
Computer Organization and Design: The Hardware/Software Interface, David Patterson and John Hennessy, Morgan Kaufmann, 3rd Edition.

Course Website: http://www.cs.uaf.edu/2010/fall/cs441

ADA Compliance: Will work with Office of Disabilities Services (203 WHIT, 474-7043) to provide reasonable accommodation to students with disabilities.

Course Goals and Requirements

By the end of the course, you will be able to understand both the present and future of computer design for performance: parallelism. Specifically, we will cover circuit-level parallelism via circuit simulators; instruction-level transparent parallelism including pipelining, multi-issue superscalar, and out-of-order execution; vector processing including SWAR, SIMD, and GPU programming; as well as coarser-grained parallelism including multicore, multi-thread, and distributed-memory. To understand this, you will need to know at least the following topics from the course prerequisites:


Last day to drop: Friday, September 17. Midterm exam: Tuesday, October 19. Last day to withdraw: Friday, October 29. Thanksgiving Break: Thursday, November 25-Sunday, November 28. Last class: Thursday, December 9. Final exam: 8am Thursday, December 16.

Student Resources

Academic Help: Google, Rasmuson Library, Academic Advising Center (509 Gruening, 474-6396), Math Lab (Chapman Room 305), English Writing Center (801 Gruening Bldg, 478-5246).


Your work will be evaluated on correctness, rationale, and insight, not on successful regurgitation of random trivia. Grades for each assignment and test may be curved up or down if needed. Your grade is then computed based on four categories of work:

  1. HW: Homeworks and machine problems, to be distributed through the semester.

  2. PROJ1: a paper and in-class presentation on an architecture topic of your choice, due in October.

  3. PROJ2: a software development or hardware performance analysis project, due in December.

  4. MT: Midterm Exam, Tuesday, October 19.

  5. FINAL: Final Exam (comprehensive), 8am Thursday, December 16.

Your overall score is then calculated as:
GRADE = 15% HW + 15% PROJ1 + 15% PROJ2+ 25% MT + 30% FINAL
This percentage score is transformed into a plus-minus letter grade via these cutoffs: A >= 93%; A- 90%; B+ 87%; B 83%; B- 80%; C+ 77%; C 70%; D+ 67%; D 63%; D- 60%; F. The grades “C-”, “F+”, and “F-” will not be given. “A+” is reserved for truly extraordinary work.

The Fine Print

At my discretion, I may round your grade up if it is near a grading boundary. Homeworks are due at midnight on the day they are due. Late homeworks will receive no credit. At my discretion, I may allow late work without penalty when due to circumstances beyond your control. Projects that are up to two weeks late may be accepted at a 50% grade penalty (e.g., on-time grade: 86%; late grade: 43%). Everything you turn in must be your own work--violations of the UAF Honor code will result in a minimum penalty equal to THAT ENTIRE SECTION OF YOUR GRADE (e.g., one plagiarized homework question will negate an otherwise perfect grade on all homeworks). However, even substantial reuse of other people's work is fine (and not plagiarism) if it is clearly cited; you'll be graded on what you've added to others' work. Group projects (NOT homeworks) are acceptable iff you clearly label who did what work; but I do expect a two-person group project to represent twice as much work as a one-person project. Department policy does not allow tests to be taken early; but in extraordinary circumstances may be taken late. In extraordinary circumstances, such as an infectious disease outbreak, classes may be held on Blackboard/Elluminate Live.

Course Outline (Tentative)

(September: 1950 through 2000 AD)

Physical Parallelism

  • Small circuit simulation with tkgate

  • Industrial circuit simulation with VHDL

  • Photolithography: PCB, semiconductor

Performance background:

  • Timing your code in NetRun

Instruction-level Parallelism

  • Pipelining

  • Operand forwarding (register file bypass)

  • Pipeline hazards and data dependencies

  • Superscalar execution (wide issue)

  • Out-of-order execution

  • Branch prediction & speculation

(October: technology post 2000AD)

Vector Parallelism

  • SWAR, SIMD, and SSE

  • SSE branch instructions & quantum superposition

GPU Programming

  • Barrel processors, MTA

  • OpenGL Shading Language (GLSL)

  • CUDA and OpenCL

  • Non-graphics code on the GPU: GPGPU

Project 1 Presentations

(Midterm To Thanksgiving)

Multicore Parallelism

  • SMP, SMT, multicore hardware

  • Shared-memory programming with threads, OpenMP, Intel TBB

  • Locks and race conditions

Speeding Up Memory

  • Cache hardware design, thrashing

  • Cache hit ratio, performance modeling

  • Cache coherence in a multicore world

  • False sharing and multicore cache thrashing

(Thanksgiving to End)

Distributed-memory Parallelism

  • Fork & mmap on multicore machine

  • Clusters, MPP, and cloud computing

  • Clients, Servers, and Peer-to-Peer

  • Network interfacing via sockets

  • MPI, the Message Passing Interface

  • glMPI and cudaMPI (Dr. Lawlor research!)

Project 2 Presentations