CS 2450
Parallel Computing
Fall 2008
Tuesdays and Thursdays from 9:30PM to 10:45 PM
5313 Sennott Square
Instructor
Rami Melhem (melhem@cs.pitt.edu)
6137 Sennott Squarre
phone: 624-8493
Office Hours:
Tuesdays and Thursdays from 10:45:00 to 12:30
Reference Books
-
Introduction to Parallel Computing, Second Edition
by Ananth Grama, Anshul Gupta, George Karypis and Vipin Kumar
Publisher: Addison-Wesley
-
Chip Multiprocessor Architecture: Techniques to Improve Throughput and Latency
by Kunle Olukotun, Lance Hammond, and James Laudon
Publisher: Morgan & Claypool
-
Parallel Computer Architecture: A Hardware-Software Approach
by David Culler and Jaswinder P. Singh, with Anoop Gupta Tanenbaum
Publisher: Morgan Kaufmann
-
Priciples of Parallel Programming
by Calvin Lin and Lawrence Snyder
Publisher: Addison-Wesley
Requirements and grading:
-
Three assignments 25%
-
Project 40%
-
Finals 35%
Course Outlines:
-
Preliminaries
- Definitions, models of computation, architecture models,
performance measures, simple parallel algorithms, ...
-
Introduction to Parallel Programing
- Message passing, shared memory, task and data parallelism, ..
-
Programing Parallel Machines
- Pthreads, MPI, OpenMP, UPC, Map-reduce, ...
-
Interconnection Networks
- Topologies, control, routing, communication efficiency and cost, ..
-
Shared Memory Multiprocessors
- UMA and NUMA architectures, snooping and directory-based cache coherence, synchronization, ...
-
Multicore Processors
- Architecture, communication, resource and power management, latency tolerance, CMP simulators, ...
-
Parallelizing algorithms
- task scheduling and mapping, load balancing,
-
Parallel Benchmarks
Flash, Parsec, ...
-
Efficient parallel algorithms
- Sorting, matrix algorithms, FFT, ...
PSC accounts:
We will be using PSC machines
(both SMP and MPP) as programming platforms.
To get an account on the PSC machines, email the following information
to Karen at karen@cs.pitt.edu.
Name
Address (Full U.S. mail address)
E-mail address