CSE 548 - Winter 2002

Computer Systems Architecture


Instructor
Mark Oskin
oskin@cs.washington.edu
Office: 325C Sieg
Office Hours: TBA

Lecture Tuesday & Thursday 5:00 - 6:20pm EE1 045
Teaching Assistant
Tim James
tjames@cs.washington.edu
Office: 233 Sieg
Office Hours: TBA

Announcements:

  • MIDTERM 2/5
  • Homeworks

    Readings and Lecture Notes

    Tuesday 1/8Introduction
    Thursday 1/10 Retrospective on High-Level Language Computer Architecture
    A VLSI RISC
    Optional: CryptoManiac: A Fast Flexible Architecture for Secure Communication"
    Tuesday 1/15 The Engineering Design of the Strech Computer
    H&P Chapters 3.1 - 3.5
    Thursday 1/17 Excerpts from Design of a Computer: the Control Data 6600
    H&P Pages 221 - 251
    Optional: Parallel Operation in the Control Data 6600, 1964
    Tuesday 1/22 An Efficient Algorithm for Exploiting Multiple Arithmetic Units
    H&P Pages 251 - 261
    Thursday 1/24 A VLIW Archicture and the ELI-512
    Alpha versus IA-64 (From DEC/Compaq's perspective)
    Optional: A VLIW Archicture -- Retrospective
    Tuesday 1/29 A Study of Branch Prediction Strategies
    Retrospective: A Study of Branch Prediction Strategies
    An Analysis of Correlation and Predictability: What Makes Two Level Branch Predictors Work
    Optional: A Language for Describing Predictors for Automatic Synthesis
    Thursday 1/31 Lockup Free Instruction Fetch/Prefetch Cache Organization
    Lockup Free Instruction Fetch/Prefetch Cache Organization - Retrospective
    Improving Direct Mapped Cache Performance by the Addition of a Small Fully-Associative Cache and Prefetch Buffers
    Improving Direct Mapped Cache Performance by the Addition of a Small Fully-Associative Cache and Prefetch Buffers - Retrospective
    Optional: H&P 5.1 - 5.5
    Tuesday 2/5 MIDTERM
    Thursday 2/7 Clock Rate Versus IPC: The End of the Road for Conventional Microarchitectures
    Tuesday 2/12 SMT Maximizing On Chip Parallelism
    SMT Maximizing On Chip Parallelism - Retrospective
    The Case for a Single Chip Multiprocessor
    Thursday 2/14 Can Dataflow Subsume Von-Neumann Computing
    Towards a Universal Building Block for Molecular and Silicon Computing
    Tuesday 2/19 System Architecture Directions for Network Sensors
    Quantifying the Energy Consumption of a Pocket Computer
    Thursday 2/21 Student Presentations:
    • Amol Prakash & Sumit Sanghai - Optimality of Tomasulo's Algorithm Presentation & Paper
    • Doug Beal & Ashish Gutpa - Virtualization in x86 Presentation & Paper
    • Andrew Schwerin & Rick Cox - Building an SMT Application Simulator Paper
    Tuesday 2/26 Student Presentations:
    • Aseem Agarwala & Antoine McNamara need title Presentation & Paper
    • Chris & Sahngyun - A Framework and Analysis of Modern Graphics Architectures for General Purpose Programming Paper
    • Miryung Kim (Kimmy) & Andrew Petersen - Increasing Confidence in Proper Execution Through Invariant Checking Presentation & Paper
    Thursday 2/27 Student Presentations:
    • Zizhen Yao & Song (Sam) Li - A Power Analysis Simulator for Alpha Architecture Presentation & Paper
    • Nick Deibel & Kevin Sikorski - Exploring Perceptrons in Branch Prediction Presentation & Paper
    • Lin Liao & Tian Sang - Optimality of Tomasulo's Algorithm Presentation & Paper
    • Luna Dong & Gang Zhao - Optimality of Tomasulo's Algorithm Presentation
    • Paper
    Tuesday 3/5 Virtual Memory: Issues of Implementation
    Virtual Memory in Contemporary Microprocessors
    H&P 5.11
    Optional: H&P 5.7 - 5.10
    Thursday 3/7 Pentuim 4
    Alpha 21264a
    Optional: Power 4
    Optional: UltraSparc III
    Tuesday 3/12 Nanofabrics: Spatial Computing using Molecular Electronics
    Quantum Wires: Architectural Implications of Quantum Data Transport in Silicon
    Quantum Wires -- Rejected review comments
    Optional: A Practical Architecture for Reliable Quantum Computers