RIT Computer Engineering

MY HOME PAGE

Friday, 18-Aug-2017 08:29:03 EDT


6 Visitors Since 15-March-1999


  

Assignment #1, Due Thursday, April 15th: PDF, Word97
   (Files for question 10: Master Program: psum.c, Slave Program: spsum.c,   rand_data.txt)

Assignment #2, Due Thursday, April 29th: PDF, Word97

  

For the following lecture notes you can download or view a lecture as an Acrobat PDF file, or as a Microsoft Powerpoint97 file:

3-9-99 The Need and Feasibility of Parallel Computing, Technology Trends, Microprocessor Performance Attributes, Goal of Parallel Computing. Computing Elements, Programming Models, Flynn's Classification, Multiprocessors Vs. Multicomputers. Current Trends In Parallel Architectures, Communication Architecture.

3-11-99 Parallel Architectures Convergence: Naming, Operations, Ordering, Replication. Communication Cost with respect to various programming models, Communication Cost Model

3-16,18-99 Parallel Programs: Conditions of Parallelism. Asymptotic Notations for Algorithm Analysis, PRAM. Levels of Parallelism, Hardware Vs. Software Concurrency. Amdahl’s Law, DOP, Concurrency Profile. Steps in Creating Parallel Programs: Decomposition, Assignment, Orchestration, Mapping.

3-23-99 Parallel Programming for Performance

3-30-99 Message-Passing Programming: Parallel Virtual Machine (PVM).

4-6-99 Message-Passing Computing Examples: Image Transformations, Mandelbrot Set, Divide-and-conquer Problem Partitioning: Parallel Bucket Sort, Numerical Integration.

4-13-99 Synchronous Iteration: Barriers, Iterative Solution of Linear Equations. Dynamic Load Balancing: Centralized, Distributed, Moore's Shortest Path Algorithm.

4-15-99 Network Requirements For Parallel Computing. Static Point-to-point Connection Network Topologies. Network Embeddings. Dynamic Connection Networks.

4-20,22-99 Shared Memory Multiprocessors. The Cache Coherence Problem. Memory Access Consistency Models. Cache Coherence Approaches. Snoopy Bus Protocols: Write-invalidate: MSI, MESI, Write-Update: Dragon.

4-27-99 Parallel System Performance: Evaluation & Scalability. Workload Selection. Parallel Performance Metrics Revisited. Application Models of Parallel Computers. Parallel System Scalability.

4-29-99 Scalable Distributed Memory Machines. MPPs Scalability Issues: Node, Network, Communication Assist, OS, Cost. Message Processing Issues: Functionality of CA, Physical DMA, System-Level Vs. User-Level Ports. MPP Physical Scaling Examples: nCUBE/2, CM-5, IBM SP-2, iWARP, Intel Paragon.

5-6-99 Cache Coherence in Scalable Distributed Memory Machines.

5-11-99 Exam Review.


  

Tuesday and Thursday: 5:00-6:50 PM, 9/2139


  

Dr. Muhammad Shaaban
e-mail: meseec@osfmail.isc.rit.edu
Office: 17-2507 X2373


Office Hours:
My Spring 99 schedule


  

Current: http://www.rit.edu/~meseec/eecc756-spring99/
Spring 98: http://www.rit.edu/~meseec/eecc756-spring98/
Spring 97: http://www.rit.edu/~meseec/eecc756/


  

This course covers a number of issues involved in the design and utilization of high performance parallel computing systems. This includes: parallel computer models, the concept of scalable performance, the memory hierarchy, cache coherence issues, parallel and scalable architectures, parallel programming concepts. A number of current parallel machines will be studied.


  

Advanced Computer Architecture EECC-722.


  

Textbooks:

Parallel Computer Architecture: A Hardware/Software Approach, David E. Culler, Jaswinder P. Singh, Morgan Kaufmann Publishers, 1999.

Parallel Programming: Techniques and Applications Using Networked Workstations and Parallel Computers, Barry Wilkinson, Micheal Allen, Prentice Hall, 1998.  ISBN: 0-13-671710-1  Buy online from Amazon.com


References:

Designing and Building Parallel Programs, Ian Foster, Addison-Wesley, 1995, complete textbook online.

PVM (Parallel Virtual Machine)

PVM: Parallel Virtual Machine: A Users' Guide and Tutorial for Networked Parallel Computing, Al Geist(Editor), et al, MIT Press, 1994, complete online version.

Advanced Tutorial on PVM 3.4 New Features and CapabilitiesAl Geist, Presented at EuroPVM-MPI'97, 1997.

Scalable Parallel Computing, Kai Hwang, Zhiwei, McGraw-Hill, 1998.

Advanced Computer Architecture: Parallelism, Scalability, Programmability, Kai Hwang, McGraw-Hill, 1993.

Selected papers.


  

Participation and class presence: 5%
Homework assignments: 30%
Exam: 35%
Class project: 30%


  

Attending all lecture sessions is expected.


  

Week 1: Parallel Computer Models
Week 2: Program and Network Properties
Week 3: Principles of Scalable Performance
Week 4: Processor and Memory Hierarchy
Week 5: Bus, Cache, and Shared Memory
Week 6: Pipelining and Superscalar Techniques
Week 7: Multiprocessors and Multicomputers
Week 8: Multivector and SIMD Computers
Week 9: Software for Parallel Programming

  This page is 20Kbytes long and was last modified on:   Monday, 25-Feb-2008 17:25:15 EST.

Made with at least 30% post-consumer recycled bits