|David E. Culler
In addition to the textbook, this course includes a number of readings from research papers. Such papers are important for a number of reasons, not the least of which is to understand that design decisions are not always black and white. Students will also undertake a major computing systems analysis and design project of their own choosing.<>This term I will try mixing a lecture format to cover the wealth of conceptual background with extensive class discussion. We have a very good text and a solid basis of original papers. You must do the reading assigned in the schedule for each class prior to the class meeting and bring one question raised by what you have read. Questions will be collected at the beginning of class. The first part ofthe class will be in lecture format, but the last third will try to answer outstanding questions.
Departmental Grading Guidelines for Graduate courses
| Professor David
627 Soda Hall, 643-7572, firstname.lastname@example.org
Office Hours: Tues 1:00 - 2:00, Fri 11-12, or by appt.
contact Willa Walker, 643-2568,
willa@EECS.Berkeley.EDU, 626 Soda, for appt.
|J. L. Hennessy and D. A. Patterson, Computer Architecture: A Quantitative Approach, 3rd Edition, Morgan Kaufmann Publishing Co., Menlo Park, CA. 2002.|
|(recommended, readings will be available on-line)
Readings in Computer Architecture, Mark Hill (Editor), Norman Jouppi (Editor), Gurindar Sohi (Editor), Morgan Kaufmann Publishing Co., Menlo Park, CA. 1999
Every effort will be made to get the notes on the web prior to the lecture. Note, however, that the notes may be updated slightly following the lecture. Click here for instructions regarding how to view pdf files.
Permission is granted to copy and distribute this material for educational purposes only, provided that the complete bibliographic citation and following credit line is included: "Copyright 2005 UCB." Permission is granted to alter and distribute this material provided that the following credit line is included: "Adapted from (complete bibliographic citation). Copyright 2005 UCB."
This material may not be copied or distributed for commercial
without express written permission of the copyright holder. The only
is for copies of these lecture notes for course readers from copy
like Copymat or Kinkos.
|1||Tu||18-Jan||Introduction (ppt, pdf)
||H&P Ch 1&2||In class debate on 1/25
pipeline, cache (ppt,pdf)
||H&P App A.1-3, ch 3.1, ch 5.1-3|
|3||Tu||25-Jan||ISA debate||IBM 360,B5000, RISC,
|4||Th||27-Jan||ControlFlow, Pipeline Traps, Banch Prediction (ppt, pdf)diction||App A.4-6, 3.4-5,4.2||HW 1, due Feb
3 start of class
|5||Tu||1-Feb||Out-of-order Completion (ppt, pdf)||3.6, 4.1, 4.2, A8, Cray-1, CDC6600
||Bring 1 question on each paper
|6||Th||3-Feb||Catch up (ppt, pdf)|
|7||Tu||8-Feb||Dynamic Inst. Scheduling (ppt, pdf)||Ch 3.2-3, Tomasulo||HW1 due|
|8||Th||10-Feb||Loop Level Parallelism (ppt, pdf)||ch 3.8-9, 4.3-4|
|9||Tu||15-Feb||Precise exceptions (ppt, pdf)
|10||Th||17-Feb||Vector Processing (ppt, pdf)
||3.7-9, 3.13,4.5-7, app G|
|11||Tu||22-Feb||Review (ppt, pdf)
|Wed||23-Feb||mid term I (solution)
|12||Th||24-Feb||Cache design (ppt, pdf)
|13||Tu||1-Mar||Shared Memory (ppt, pdf)||6.1-4, MPR handout
|14||Th||3-Mar||SMPs, Embedded Proc
||ch 6.9,6.12, SMT,
||ch 6.5-6, 6.11, scaling
Field Trip to NERSC - Bill Kramer
arch, deep data
Synchronization [ppt, pdf]
||ch 6.7-8, tutorial|
|20||Th||31-Mar||Power Management in
Wireless Sensor Networks
||Project Topic Finalized
|21||Tu||5-Apr||Cray-3 (Seymore video)
|23||Tu||12-Apr||Storage Technology and
I/O [ppt, pdf]
||Project Annotated Bibliography
|Wed||13-Apr||mid term II (poster 4/17, due 4/27)
|24||Th||14-Apr||network processors||Spalink SOSP, Crowley ICS, Kounavis|
|25||Tu||19-Apr||low power design||Low-power CMOS, Variable-Voltage Core-Based Systems|
||Project Preliminary Results
|28||Th||28-Apr||no class, faculty
|29||Tu||3-May||Whatever happened to
Retrospective (if you
|30||Th||5-May||Seth Goldstein: Claytronics
-realizing programmable matter
More info from Westley Weimer: "The web requests (the only ones I have examined in detail) tend to be long text files of tuples: (client-ip, time, document requested, size, etc) with the exact fields and format varying by trace. One trace even comes with a C interface for reading their records. "
For a precompiled version that works on Linux X86, see Kubiatowicz's home page: ~kubitron/simplescalar.