Login
Course Information
Identification

CS 421: Parallel Programming Paradigms
 
Description

Motivation for parallel computing

Current many-core, multi-core, heterogeneous computers: architecture

Analytical modeling of parallel programs

Basics of current programming models for computation in parallel with C or C++ (examples: pthreads, OpenMP, MPI, CUDA)

Basics of Java based parallel computing and Cloud Computing: (examples: Java 8 Map Reduce, Java Streams)

Implementation of classic and instructive algorithms using MPI,

OpenMP, and CUDA (examples: prefix sums, sorting, matrix multiplication, random numbers, image filtering, convex hulls)

Implementation of selected scientific algorithms using MPI, OpenMP, and CUDA (examples: differential equations, numerical linear algebra, Monte-carlo methods, fast fourier transform)

Implementation of selected graph algorithms using MPI, OpenMP, and CUDA (examples: minimum spanning tree, shortest paths, search)

Selected Industry or Government case studies (examples: Multithreading for
Visual Effects, Computational Biology, Molecular Dynamics, Big Data)

Central concepts in selected topics: Use of GPUs for Deep Learning, Automatic parallelization, Bulk Synchronous Parallel Abstract Machine
 
References

Will be provided on web page.
 
Home Page

~sharat/current/cs421/
 
Prerequisites

CS101, CS213 or equivalent.Courses that are not enforced but exposure required: CS 347 or equivalent, CS 305 or equivalent
 
Other Details

Duration : Full Semester Total Credit : 6
Type : Theory
 
Current Semester (Autumn 2017-18)

Status : Not Offered Instructor : ---
 
Next Semester (Spring 2017-18)

Status : Offered Instructor : Prof. Sharat Chandran




Last Modified Date: 09-May-2016

Webmail

Username:
Password:
Faculty CSE IT
Forgot Password
    [+] Sitemap     Feedback