Course Information for CS252: Graduate Computer Architecture
Spring 2018
Catalog Description: Graduate Computer Architecture
4.0 units. Three hours of lecture plus one hour paper discussion per
week.
Prerequisites: CS152
or equivalent.
Class Schedule/Rooms
Lectures: Monday and Wednesday, 2:30-4:00pm, 306 Soda Hall
Paper discussion: Monday, 4-5pm, 405 Soda Hall
Both midterms held in class during lecture times.
Instructor: Krste
Asanović, Professor, CS Division, EECS Department
Email: krste at eecs
Office Hours: Wednesday 9:30-10:30am, 579 Soda Hall (email to confirm)
TA: Donggyu Kim
Email: dgkim at eecs
Office Hours: Thursdays 2-3pm, 611 Soda Hall
TA: Howard Mao
Email: zhemao at eecs
Office Hours: Tuesdays 10-11am, 611 Soda Hall
Course Grading
15% Problem Sets
We will distribute 5 problem sets for you to practice your
understanding of the course material. The problem sets also provide
essential background material for the mideterms. The problem sets will
be graded primarily on an effort basis, but if you do not work through
the problem sets you are unlikely to succeed at the midterms! We will
distribute solutions to the problem sets soon after the problem sets
are due to give you instant feedback.
15% Paper readings
Paper summaries are posted as private notes on the Piazza thread
for each paper. Each summary should include one paragraph
summarizing main content of paper including good/bad points,
plus 1-3 questions to contribute to paper discussion. We will
not count the two lowest weeks' grades, which includes absence,
but please send in summaries even if you cannot attend class.
30% Exams
There will be 2 midterms (15% for each) covering the class
material. These will be closed book with no calculators,
phones, smart watches, or computers allowed.
40% Class Project
Substanial research project by pairs of students, with 10-page
conference-style paper and class presentation.
Late Assignment Policy
Problem sets must be handed in at the beginning of class on the
due date, with no extensions possible. No other extensions will
be given, unless for serious documented emergencies. An
automatic F grade is given if the course project is not
completed.
Collaboration Policy
The problem sets are intended to help you learn the material, and we
encourage you to collaborate with other students and to ask questions
in discussion sections and office hours to understand the problems.
However, each student must turn in their own solutions to the
problems.
Regrade Policy
For addition errors in the total score, return the midterm back to the TA
to get it fixed. For regrades, return the midterm to the TA within a week
of the midterm being graded with a separate sheet of paper explaining the
discrepancy. The staff will carefully regrade the entire midterm, read
the reasoning provided, and then make a final decision. Since the
entire midterm is being regraded, it is possible the total score could go
down as a consequence of previously undiscovered mistakes being
found. We therefore recommend that regrade requests only be used when
the case is strong and a significant number of points are at stake.
Piazza
The course will use Piazza for class communication. The course page can be
found at piazza.com/berkeley/spring2018/cs252.
Compute Resources
All labs will be available on {icluster6, icluster7, icluster8, icluster9}.eecs. You may want to use other
servers for writing code or compiling, and you can see a full list of them at
inst.
Textbooks
The following textbook is highly recommended for the course:
|
J. L. Hennessy and D. A. Patterson, Computer
Architecture: A Quantitative Approach, 6th Edition, Morgan
Kaufmann Publishing Co., Menlo Park, CA. December 2017.
ISBN13: 978-0128119051
ISBN10: 0128119055
We will also use material from
the companion
Web site.
|
The following textbook is recommended to refresh your background and
to provide a simpler introduction to some of the basic concepts. Any
recent edition should be sufficient for background study.
|
D. A. Patterson and J. L. Hennessy, Computer
Organization and Design RISC-V Edition: The Hardware Software Interface,
1st Edition, Morgan Kaufmann Publishing Co., Menlo Park, CA., April 2017.
ISBN13: 978-0128122754
ISBN10: 0128122757
|