ECE1755 Fall 2007:
Parallel Computer Architecture and Programming

Home Administrivia Lectures/Readings/Reviews/Events Projects CCnet/BBoard

Lectures and Other Events

Future events are tentative.
Lecture slides are available in pdf form (requiring acrobat reader).

Phase 1

Date Lecture Readings (due Monday night before corresponding class) Homework/Project Out Homework/Project Due
Sep 12 Lecture 1: Past, Present, and Future of Computing ( pdf, 2up.pdf, 4up.pdf ); Administration ( pdf, 2up.pdf, 4up.pdf ); Background ( pdf, 2up.pdf, 4up.pdf ); Homework1 out
Sep 19 (no class) Review: Olukotun; CS 1;
Sep 26 Lecture 2: Parallel Programming ( pdf, 2up.pdf, 4up.pdf ) Review: Woo; CS 2,3;
Oct 3 Lecture 3: Interconnects and Coherence (pdf, 2up.pdf, 4up.pdf ) Review: Stenstrom CS 5,7.1-2,10.1-4 Homework2 out Homework1 in
Oct 10 Lecture 4: Scalable Coherence and Consistency (pdf, 2up.pdf, 4up.pdf ) Review: Gharachorloo; CS 8.1-4, 9.1
CS = Parallel Computer Architecture---A Hardware/Software Approach, by Culler, Singh, and Gupta (recommended textbook, )

Phase 2

Date Lecture Readings (due Monday night before corresponding class) Presenters Homework/Project Due
Oct 17 Multicore Organization Review:
1: Exploring the Design Space of Future CMPs;
2: Maximizing CMP Throughput with Mediocre Cores;
3: Single-ISA Heterogeneous Multi-Core Architectures for Multithreaded Workload Performance; 		1: Kaveh;
2: Eric;
3: Cedomir 		Homework2 in
Oct 24 Multicore Sharing Review:
1: Computation Spreading;
2: Cooperative Caching for Chip Multiprocessors;
3: CMP Phase Co-Scheduling; 		1: Keith;
2: Gary;
3: Somu 		Project email proposal in
Oct 31 Helper Threads Review:
1: Physical Experimentation with Prefetching Helper Threads on Intels Hyper-Threaded Processors;
2: Speculative Precomputation on Chip Multiprocessors;
3: Dynamic Helper Threaded Prefetching on the Sun UltraSPARC CMP Processor; 		1: Paul;
2: Thomas;
3: Bernice 		Project written proposal in
Nov 7 Thread-Level Speculation Review:
1: A Scalable Approach to Thread-Level Speculation;
2: Tolerating Dependences Between Large Speculative Threads Via Sub-Threads;
3: Program Demultiplexing; 		1: Stan;
2: David Han;
3: Alireza
Nov 14 Transactional Memory (TM) and Software TM (STM) Review:
1: The Atomos Transactional Programming Language;
2: JudoSTM: A Dynamic Binary-Rewriting Approach to Software Transactional Memory;
3: Architectural Support for Software Transactional Memory; 		1: Diego;
2: Greg;
3: James
Nov 21 Hardware TM (HTM) and Hybrid TM (HyTM) Review:
1: Transactional Coherence and Consistency;
2: Programming with Transactional Coherence and Consistency;
3: Hybrid Transactional Memory; 		1: Daniel;
2: Ivan;
3: Danny 		Project email interim in
Nov 28 Sequential to Parallel Review:
1: Revisiting the Sequential Programming Model for Multi-Core;
2: Decoupled Software Pipelining with the Synchronization Array;
3: Extending Multicore Architectures to Exploit Hybrid Parallelism in Single-Thread Applications; 		1: Adrian;
2: Dimitris;
3: Jeff
Dec 5 Domain Computing Review:
1: Tartan: Evaluating Spatial Computation for Whole Program Execution;
2: Anton, a special-purpose machine for molecular dynamics simulation;
3: ParallAX: An Architecture for Real-Time Physics; 		1: David Woods;
2: Bogdan;
3: Dmitry
Dec12 Project Presentations project presentations (2:00-5pm)
Dec21 final written report due