15-740: Computer ArchitectureFall 2007Syllabus1 Course Details at a GlanceLectures: Tuesdays, 12:00-1:20pm, NSH 1305Wednesdays, 3:00-4:20pm, NSH 3002Thursdays, , 12:00-1:20pm, NSH 1305Instructor: Todd C. Mowry, WeH 8105, 268-3725, [email protected] Hours: Mondays, 4-5pm, or by appointmentTA: Tunji Ruwase, WeH 8102, 268-3042, [email protected] Hours: by appointmentClass Administrator: Jennifer Landefeld, WeH 8120, [email protected] Page: www.cs.cmu.edu/afs/cs/academic/class/15740-f07/www/Newsgroup: cmu.cs.class.cs740Course Materials: /afs/cs.cmu.edu/academic/class/15740-f07/public2 TextbooksMain Text: Hennessy, J. L, and Patterson, D. A., Computer Architecture: A Quantitative Approach,4th Edition. Morgan Kaufmann, 2006. (NOTE: don’t try to get by with earlier editions of the book;the fourth edition is much different and more aligned to the content of this course.)Supplemental Text: Culler, D., and Singh, J.P., with Gupta, A. Parallel Computer Architecture: AHardware/Software Approach, Morgan Kaufmann, 1998. (Copies should be available in the E&SLibrary by the time we need them.)3 Course Overview and ObjectivesThis course attempts to provide a deep understanding of the issues and challenges involved in designingand implementing modern computer systems. Our primary goal is to help students become more skilled intheir use of computer systems, including the development of applications and system software. Users canbenefit greatly from understanding how computer systems work, including their strengths and weaknesses.This is particularly true in developing applications where performance is an issue.In addition to covering issues that are relevant in systems with only a single process or, the courseplaces a strong emphasis on parallel systems comprised of multiple processors, with topics ranging fromprogramming models to hardware realizations. While the Hennessy and Patterson textbook covers nearlyall of the topics that will be discussed in this course, the Culler, Singh and Gupta textbook goes intogreater depth on parallel architecture and programming.13.1 Course ThemesAn addition to our “user-centric” (vs. “builder-centric”) approach, the course has several other themes.One theme is to emphasize the role of evolving technology in setting the directions for future computersystems. Computer systems, more than any other field of computer science, has had to cope with thechallenges of exploiting the rapid advances in hardware technology. Hardware that is either technologicallyinfeasible or prohibitively expensive in one decade, such as bitmapped full color displays or gigabyte diskdrives, becomes consumer products in the next. Technology that seems to have a bright future, suchas magnetic bubble memories, never becomes competitive. Others, such as CMOS, move from being aniche technology to becoming dominant. In addition, computer systems must evolve to support changesin software technology, including advances in languages and compilers, operating systems, as well aschanging application requirements. Rather than teaching a set of facts about current (but soon obsolete)technology, we therefore stress general principles that can track evolving technology.Another theme of the course is that “hands-on” exercises generally provide more insight regardingsystem behavior than paper-and-pencil exercises. Hence our assignments involve programming and usingcomputer systems, although in a variety of different ways.Finally, rather than stopping with state-of-the-art in computer architecture as of a decade ago, anothertheme of this course is looking at the state-of-the-art today as well as open research problems that arelikely to shape systems in the future. Hence we will b e discussing recent papers on architecture researchin class, and students will perform a significant research project.4 PrerequisitesThis course is not intended to be your first course on computer architecture or organization; it is gearedtoward students who have already had such a course as undergraduates. For example, we expect thatpeople are already at least somewhat familiar with assembly language programming, pipelining, andmemory hierarchies. If you have not had such a course already, then it is still poss ible to take thiscourse provided that you are willing to spend some additional time catching up on your own. If you feeluncertain about whether you have adequate preparation, please discuss this with the instructor.In addition to an undergraduate computer organization course, here are some other topics which arehelpful for this course (references are included for self study):• A working knowledge of C, including pointers and memory allocation [Kernighan+88].• An introductory compiler course, esp ec ially aspects of code generation such as data formats, pro-cedure linkages, and translation of control constructs [Aho+86, Chs. 1–2, 7–9].• An introductory op erating system course, especially aspects of protection, scheduling, and concur-rency [Silberschatz+91, Chs. 4.1–4.3, 5.1–5.4, 5.7, and 7–9]• Bit-level representation of and manipulation of numbers [HenPat96, App. A.1–A.2].5 If You Are Not a CS or ECE PhD StudentIf you are not a graduate student in either the CS or ECE PhD program, you need permission to takethis class. If you have not already done so, send a message to the instructor stating your status, why youwant to take the class, and if you want to take the class for credit or as an auditor.6 Course WorkGrades will be based on homeworks, a rese arch project, two exams, and class participation.2Homeworks: There will be two homework assignments. Each assignment involves a non-trivial amountof programming. You will work in groups of two or three (preferably three) people on the assign-ments. (Turn in a single writeup per group.)Project: A major focus of this course is the project. We prefer that you work in groups of two on theproject, although groups of up to three may be permitted depending on the scale of project (ask theinstructor for permission before forming a group of three). The project is intended to be a scaled-down version of a real research project. The project must involve an experimental component—i.e.it is not simply a paper and pencil exercise. We encourage you to come up with your own topicfor your project, although we can give you suggestions if you are stuck. You will have six weeksto work on the project. You will present your findings in a written rep ort
View Full Document
Unlocking...