Please note that this session will be given a lecture in Japanese.
In this training seminar, we will explain the basics and how to use parallel programming using the automatic parallelization function that can be used in general OpenMP, MPI, and Vector processor as parallel programming methods. It comes with a "free trial account" that allows you to use the supercomputer freely for one week.
What is OpenMP?
This is a method of intra-node parallelization (shared memory parallelization) by inserting a line of instructions to the compiler in the program. Since parallelization is possible only by inserting instruction lines, there is an advantage that the burden of programming is less, and portability is high. However, since parallelization across nodes cannot be performed, programming in combination with MPI is necessary to perform large-scale parallelization. The description method is slightly different in C and Fortran.
What is MPI?
MPI is a de facto standard specification of message passing for distributed memory parallel processing. MPI allows users to perform computations using multiple computing nodes. Therefore, users can benefit from a large amount of memory space in total. However, as users themselves must explicitly write internode communications and split computation, a burden for programming will be heavy.
Highly recommended for…
- Those who want to speed up/parallelize your program but don't know how to do it
- Those who have a basic knowledge of Unix (who can use basic commands and editors such as emacs and vi)
- Those who have C, Fortran programming experience
Important notice regarding taking this course
This workshop is live-streamed online using online meeting tools. Please make sure you have an internet access.
The materials for the workshop
Date : June 23, 10:00 a.m - 16:00 p.m.
Instructor: NEC Corporation
Type : Lecture、Hands-on
Application deadline: June 22, noon
We stopped accepting applications because there is no more place available.