Mahadevan Gomathisankaran

Course Info

• My Office: F227 Discovery Park
• My Phone: 940-565-4864
• My Email: mgomathi@unt.edu
• TA:  Shyamala Balasubramanian
• Office Hours: Tu Th 2 - 3 PM and by appointment
• Class Hours: Tu Th 3:30 - 4:50 PM @ NTDP B158

Announcements

• 02/16/10 - Mid term exam scheduled for 03/11/2010.
• 02/09/10 - Assignment 1 Graded.
• 01/22/10 - Assignment 1 posted. Due on 01/29/2010 11:30 PM.

Grade Distribution

Syllabus

The purpose of this course is to provide you with a solid foundation in computer systems architecture. This course is generally considered as a foundation to further study and research in computer systems. We will survey several different approaches to designing a single CPU that can aid in building a parallel processor. We will investigate instruction level parallelism, branch prediction techniques, various cache organization, multithreaded architectures, cache coherency and their impact on parallel processing.

Course Outline:

• Introduction and Background
• • What is computer architecture
  • Instruction sets, control unit, etc.
  • Performance evaluation
  • Qunatitative analysis
  • Basic cryptographic protocols
• Pipelining
• • Basic design of pipelines
  • Data and control hazards
  • Branch prediction and dynamic scheduling
  • Superscalar and multiple issue processors
  • VLIW or EPIC
• Dataflow and multithreaded architectures
• • Dataflow model of computation
  • What is multithreading
  • Scheduled Dataflow
  • SMT and Hyper Threading
• Cache Memories
• • Cache memory designs
  • Improving cache performance
  • Cache coherency
• Shared Memory Multiprocessors
• • Mutual Exclusion and Synchronization
  • Cache Coherency Problem and Solutions
  • Distributed Shared Memory Systems
• Low Power Issues

References

• Textbook: J. Hennessy and D. Patterson. “Computer Architecture: A Quantitative Approach”, 3rd Edition (or 4th edition)
• J.P. Shen and M.H. Lipasti. “Modern Processor Design”, McGraw Hill, 2005
• D.E. Culler and J.P. Singh. “Parallel Computer Architecture”, Morgan Kaufman, 1999

Prerequisites

• Computer Organization (CSCE 2610 or CSCI 3100) and Systems Programming (CSCIE3600). Primarily, I would like to see that you know how a basic CPU works (including instruction fetch, decode, execute cycles, microprogramming), instruction sets and choices (including RISC versus CISC, address modes), memory organization (including virtual memory, memory interleaving), some I/O (including interrupts, polling, daisy chaining, serial and parallel I/O), ALU (including multiplication, division and floating point algorithms), synchronization and mutual exclusion (like semaphores), some understanding of networks and protocols, some understanding of compilers and runtime support, Operating system concepts like process scheduling, virtual memory, protection domains, etc.

Policies

Academic Dishonesty:

Cheating in exams/assignments, plagiarism in exams/assignments, collusion and falsification of academic records or the attempt to do these things constitute academic dishonesty. Students need to include proper citation for books and/or Internet based resources on their submissions. All submissions will be checked against plagiarism via SafeAssign tool. Any direct copying from a book, from an Internet site, from a paper, etc. without proper citation to the author (of the book or the Internet article) is considered as plagiarism. All assignments need to be done individually. Any type of academic dishonesty will be handled immediately and strictly, resulting in a zero on the exam/assignment and an F in the class, and will be referred to the Dean for further disciplinary action.

Students with Disabilities:

Students in need of academic accommodations for disability can refer to the UNT Policy Manual for initiating the required arrangements based on ADA terms. Also, disabled students can arrange an appointment with me to discuss their special needs for academic accommodation during office hours (to ensure confidentiality).

Observation of Religious Holidays:

In accordance with Texas state law, a student absent due to the observance of a religious holiday may take examinations or complete assignments scheduled for the days missed, including those missed for travel, within a reasonable time after the absence. Students should notify the instructor in each course of the date of the anticipated absence as early in the semester as possible. Only holidays or holy days observed by a religion whose place of worship is exempt from property taxation under Section 11.20 of the Tax Code may be included. A student who is excused under this provision may not be penalized for the absence, but the instructor may appropriately respond if the student fails satisfactorily to complete the assignment or examination.

Student Evaluation:

The Student Evaluation of Teaching Effectiveness (SETE) is a requirement for all organized classes at UNT. This short survey will be made available to you at the end of the semester, providing you a chance to comment on how this class is taught. I am very interested in the feedback I get from students, as I work to continually improve my teaching. I consider the SETE to be an important part of your participation in this class.