Introduction to Parallel Programming for Multicore/Manycore Clusters
February 22-24, 2018
National Taiwan University, Taipei, Taiwan

Instructors:
  Takahiro Katagiri (Information Technology Center, Nagoya University, Japan), and
  Kengo Nakajima (Information Technology Center, The University of Tokyo, Tokyo, Japan)

Fundamental (2/22-23) (Instructor: Takahiro Katagiri )

Trainings of fundamental MPI and OpenMP are provided with parallelization of dense matrix-vector multiplications and power method for eigenvalue problems.


Fundamental 1/2: February 22th, 2018 (Thursday)

PDF / Source Codes Time Contents 

Overview_Of_OpenMP.pdf

10:10-11:00 Overview of OpenMP

Functions_Of_OpenMP.pdf

11:10-12:00 Functions of OpenMP

Overview_Of_MPI.pdf

13:10-14:00 Overview of MPI


14:10-15:00 How to use the Reedbush-U (by Prof. Nakajima)

Trainings_On_the_Reedbush-U.pdf

15:10-16:00 Trainings on the Reedbush-U

Trainings_On_OpenMP.pdf

16:10-17:00 Trainings on OpenMP


  Exercise 1
(OpenMP parallelization with the Reedbush-U)


Fundamental 2/2: February 23rd, 2018 (Friday)

PDF / Source Codes Time Contents 

Non-blocking_Comm.pdf

09:10-10:00 Functions of MPI Non-blocking and Persistent Communication

Dense_Mat-Vec.pdf

10:10-11:00 Parallelization of dense Matrix-Vector Multiplications (1/2)


11:10-12:00 Parallelization of dense Matrix-Vector Multiplications (2/2)
    Exercise 2
(Parallelization of dense Matrix-Vector Multiplications with the Reedbush-U)

Power_Method.pdf

13:10-14:00 Parallelization of dense Power Method for eigenvalue problem (1/2)

14:10-15:00 Parallelization of dense Power Method for eigenvalue problem (2/2)
Exercise 3
(MPI Parallelization of dense Power Method for eigenvalue problem with the Reedbush-U)

Mat-Mat_1.pdf

15:10-16:00 Parallelization of dense Matrix-Matrix Multiplication


Exercise 4
(MPI Parallelization of dense Matrix-Matrix Multiplication with the Reedbush-U)

Mat-Mat_2.pdf

16:10-17:00 Parallelization of Fully Distributed dense Matrix-Matrix Multiplication


Exercise 5
(MPI Parallelization of Fully Distributed dense Matrix-Matrix Multiplication with the Reedbush-U)


Update: 10rd/Feb./2018