02 programming with openmp 3 0 rev1.1.3, Thesis of Accelerator Physics

Download 02 programming with openmp 3 0 rev1.1.3 and more Thesis Accelerator Physics in PDF only on Docsity!

Programming with OpenMP*

Intel Software College

Copyright © 2008, Intel Corporation. All rights reserved.

Objectives

Upon completion of this module you will be able to use OpenMP

to:

• (^) implement data parallelism
• (^) implement task parallelism

Copyright © 2008, Intel Corporation. All rights reserved.

What Is OpenMP?

Portable, shared-memory threading API

– Fortran, C, and C++

– Multi-vendor support for both Linux and Windows

Standardizes task & loop-level parallelism

Supports coarse-grained parallelism

Combines serial and parallel code in single

source

Standardizes ~ 20 years of compiler-directed

threading experience

http://www.openmp.org

Current spec is OpenMP 3.

318 Pages

(combined C/C++ and Fortran)

Copyright © 2008, Intel Corporation. All rights reserved.

Programming Model

Fork-Join Parallelism :

• Master thread^ spawns a^ team of threads^ as needed
• Parallelism is added incrementally: that is, the sequential

program evolves into a parallel program

Parallel Regions

Master

Thread

Copyright © 2008, Intel Corporation. All rights reserved.

Agenda

What is OpenMP? Parallel regions Worksharing Data environment Synchronization Curriculum Placement Optional Advanced topics

Copyright © 2008, Intel Corporation. All rights reserved.

Most OpenMP constructs apply to structured blocks

• (^) Structured block: a block with one point of entry at the

top and one point of exit at the bottom

• (^) The only “branches” allowed are STOP statements in

Fortran and exit() in C/C++

A structured block Not a structured block

Parallel Region & Structured Blocks

(C/C++)

if (go_now()) goto more; #pragma omp parallel { int id = omp_get_thread_num(); more: res[id] = do_big_job(id); if (conv (res[id]) goto done; goto more; } done: if (!really_done()) goto more; #pragma omp parallel { int id = omp_get_thread_num(); more: res[id] = do_big_job (id); if (conv (res[id]) goto more; } printf (“All done\n”);

Copyright © 2008, Intel Corporation. All rights reserved.

Agenda

What is OpenMP? Parallel regions Worksharing – Parallel For Data environment Synchronization Curriculum Placement Optional Advanced topics

Copyright © 2008, Intel Corporation. All rights reserved.

Worksharing

Worksharing is the general term used in OpenMP to

describe distribution of work across threads.

Three examples of worksharing in OpenMP are:

• omp for construct

• omp sections construct

• omp task construct

Automatically divides work

among threads

Copyright © 2008, Intel Corporation. All rights reserved.

Combining constructs

These two code segments are equivalent

#pragma omp parallel { #pragma omp for for (i=0;i< MAX; i++) { res[i] = huge(); } } #pragma omp parallel for for (i=0;i< MAX; i++) { res[i] = huge(); }

Copyright © 2008, Intel Corporation. All rights reserved.

The Private Clause

Reproduces the variable for each task

• (^) Variables are un-initialized; C++ object is default constructed
• (^) Any value external to the parallel region is undefined void work(float c, int N) {** float x, y; int i; #pragma omp parallel for private(x,y) for(i=0; i<N; i++) { x = a[i]; y = b[i]; c[i] = x + y; } }

Copyright © 2008, Intel Corporation. All rights reserved.

The schedule clause

The schedule clause affects how loop iterations are mapped onto threads schedule(static [,chunk])

• Blocks of iterations of size “chunk” to threads
• (^) Round robin distribution
• (^) Low overhead, may cause load imbalance schedule(dynamic[,chunk])
• (^) Threads grab “chunk” iterations
• When done with iterations, thread requests next set
• (^) Higher threading overhead, can reduce load imbalance schedule(guided[,chunk])
• (^) Dynamic schedule starting with large block
• (^) Size of the blocks shrink; no smaller than “chunk”

Copyright © 2008, Intel Corporation. All rights reserved.

Schedule Clause Example

#pragma omp parallel for schedule (static, 8) for( int i = start; i <= end; i += 2 ) { if ( TestForPrime(i) ) gPrimesFound++; }

Iterations are divided into chunks of 8

• (^) If start = 3, then first chunk is i ={3,5,7,9,11,13,15,17}

Copyright © 2008, Intel Corporation. All rights reserved.

Agenda

What is OpenMP? Parallel regions Worksharing – Parallel Sections Data environment Synchronization Curriculum Placement Optional Advanced topics

Copyright © 2008, Intel Corporation. All rights reserved. Task Decomposition a = alice(); b = bob(); s = boss(a, b); c = cy(); printf ("%6.2f\n", bigboss(s,c));

alice,bob, and cy

can be computed

in parallel

alice bob

boss

bigboss

cy