lec22_parallel1

lec22_parallel1 - General Introduction to Parallel...

Info iconThis preview shows pages 1–5. Sign up to view the full content.

View Full Document Right Arrow Icon
University of Mississippi Dept. of Physics and Astronomy Phys 503, Dr. Gladden General Introduction to Parallel Processing Model for parallelization (hardware) Memory architectures Programming models HW09 due today. Reading in handout. No more homework for the semester – WORK ON PROJECTS!
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
University of Mississippi Dept. of Physics and Astronomy Phys 503, Dr. Gladden Parallel computing is a very broad term describing schemes by which to break up large problems into multiple smaller problems. Some problems are easy to cast in a parallel form: Need to fit experimental data to a model at 100 different temperature points. Have 5 different machines work on 20 different data sets (temperatures) at the same time. Important characteristic: each job is independent of the results of the previous jobs.
Background image of page 2
University of Mississippi Dept. of Physics and Astronomy Phys 503, Dr. Gladden Other problems are more difficult to parallelize Molecular dynamics: each time step in the simulation depends on the state at the previous time step. Break up by space – have different CPUs work on the same time step, but different sets of atoms. The ‘boundary’ atoms are tricky! “I know how to make 4 horses pull a cart. I don’t know how to make 1024 chickens do it!” ~~Enrico Clementi
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Dept. of Physics and Astronomy Phys 503, Dr. Gladden Traditional: Serial computing instructions and data are streamed to CPU in sequence. Parallel: Problem is compartmentalized. A series of instructions are generated for each part and sent to multiple CPUs. Results are recombined for the overall solution. Seriously parallel problems Climate models, molecular dynamics, signal processing, fluid dynamics. Written in compiled languages (C, C++, Fortran,…)
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 16

lec22_parallel1 - General Introduction to Parallel...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online