Find this at http://www.npac.syr.edu/users/gcf/dbppindod/dbpp3mpi/

Designing and Building Parallel Programs 3: MPI Message Passing System

Day 1 Introduction to Parallel Programming The OpenMP Programming Language

Day 2 Introduction to MPI Sending and receiving messages Advanced features Examples The PETSc Library

Table of Contents for Designing and Building Parallel Programs 3: MPI Message Passing System

001 Designing and Building Parallel Programs
002 Outline
003 Message Passing Interface (MPI)
004 What is MPI?
005 Compiling and Linking  (in MPICH)
006 Running MPI Programs (in MPICH)
007 Sending/Receiving Messages: Issues
008 What Gets Sent: The Buffer
009 Generalizing the Buffer in MPI
010 Advantages of Datatypes
011 To Whom It Gets Sent: Process Identifiers
012 Generalizing the Process Identifier in MPI
013 How It Is Identified:  Message Tags
014 Sample Program using Library
015 Correct Execution
016 Incorrect Execution
017 What Happened?
018 Solution to the Tag Problem
019 MPI Basic Send/Receive
020 Six-Function MPI
021 Simple Fortran Example
022 Simple Fortran Example (2)
023 Simple Fortran Example (3)
024 Advanced Features in MPI
025 Collective Communication
026 Synchronization
027 Data Movement (1)
028 Data Movement (2)
029 Collective Computation Patterns
030 List of Collective Routines
031 Example: Performing a Sum
032 Buffering Issues
033 Avoiding Buffering Costs
034 Combining Blocking and  Send Modes
035 Connecting Programs Together
036 Connecting Programs via Intercommunicators
037 Regular (Cartesian) Grids
038 Regular Grid Example: Getting the Decomposition
039 Regular Grid Example: Conclusion
040 Designing MPI Programs
041 Jacobi Iteration: The Problem
042 Jacobi Iteration: MPI Program Design
043 Jacobi Iteration: MPI Program Design
044 Jacobi Iteration: MPI Program
045 Jacobi Iteration: MPI Prog. II