CS/ECE 752 Advanced Computer Architecture I Fall 2008 Section 1 Instructor David A. Wood and T. A. Khai Tran URL: http://www.cs.wisc.edu/~david/courses/cs752/Fall2008/

Homework 1 // Due at Monday Sept 22

H&P is John L. Hennessy and David A. Patterson, Computer Architecture: A Quantitative Approach, Morgan Kaufmann Publishers, Fourth Edition, 2007.

Disclaimer: To avoid being too much work, homeworks only sample course material. Exams also sample course material and will include questions on material not covered in homeworks. Homeworks will tend to cover material related to the text, reviews focus on papers, and exams on both.

Problem 1 (20 points)

H&P Problem 1.3 (page 55) with the following changes: the defect rate for Niagara is .65 per cm^2.

Problem 2 (15 points)

H&P Problem 1.4 (page 57) with the following substitions:

• Efficiency of 85%
• Idle time of 65%
• No substitution

Problem 3 (20 points)

H&P Problem 1.12 (page 61)

Problem 4 (10 points)

a) On average, the UW Computer Sciences department graduates approximately 20 PhD students a year. Assuming that there are, on average, 120 PhD students in the department (and no one quits school, etc.), how long does the average PhD student take to complete their degree?

b) Assume that a Computer Sciences PhD student spends 3 years taking classes, half of which are completely useless ( :-) ), and the remaining years doing research. If you could avoid taking the useless classes, how much could you speed up your time to degree?

c) Repeat part b) assuming you avoid sleeping, eating, laundry, and hanging out at Memorial Union Terrace.

Problem 5 (15 points)

The Intel Speedstep and Transmeta Longrun technology promise to allow a user (or operating system) to reduce power dissipation by an order of magnitude, yet reduce performance by only about a factor of 3. Explain how this is possible.

Problem 6 (20 points)

a) In two-three paragraphs, contrast the EDGE ISA and the MIPS ISA. Your comparison should address: instruction set length, operand types, memory addressing modes, immediate types, and any special features (like special instructions, explicit or implicit condition codes). You may need to do a little additional reading to answer this well.

b) Colwell, et al. make a number of arguments about what makes a "good" ISA. In two-three paragraphs, discuss how these arguments apply to the EDGE ISA.