Différences

Ci-dessous, les différences entre deux révisions de la page.

Lien vers cette vue comparative

Les deux révisions précédentes Révision précédente
Prochaine révision
Révision précédente
intel-cluster-studio [2014/04/25 07:12] – [Compilers Options] adminintel-cluster-studio [2015/02/27 13:18] (Version actuelle) – supprimée admin
Ligne 1: Ligne 1:
-====== Intel Cluster Studio XE 2013 ====== 
  
-Intel Cluster Studio XE 2013 meets the challenges facing HPC developers by providing, for the first time, a comprehensive suite of tools that enables developers to boost HPC application performance and reliability. It combines Intel’s proven cluster tools with Intel’s advanced threading/memory correctness analysis and performance profiling tools to enable scaling application development for today’s and tomorrow’s HPC cluster systems. It contains: 
- 
- 
-  * Intel® C++ Compiler XE 14.0.2 
-  * Intel® Fortran Compiler XE 14.0.2 
-  * Intel® Debugger 13.0 Update 1 (for Linux* OS only) 
-  * GNU* Project Debugger (GDB*) 7.5 
-  * Intel® Integrated Performance Primitives 8.0 Update 1 
-  * Intel® Threading Building Blocks (TBB) 4.2 
-  * Intel® Math Kernel Library (MKL) 11.1 
-  * Intel® MPI Library 4.1 Update 1 
-  * Intel® Trace Analyzer and Collector 8.1 Update 3 
-     
-====== Compiling Applications ====== 
- 
-<note>Intel cluster suite is only installed on Lumière cluster.</note> 
- 
-Before you start compiling your application, you need to load the corresponding module: 
- 
-<code> 
- 
-module load intel/14.0.2 
- 
-</code>  
- 
- 
-**Sequential programs** : ''ifort'', ''icc'', ''icpc'' 
- 
- 
-<del>**Parallel programs (MPI) **: ''mpiifort,  mpiicc, mpiicpp''</del>  
- 
-Compiler command line: 
- 
-<code bash> 
-$ icc [options] myprogram.c -o myprogram  
-</code> 
- 
-===== Compilers Options ===== 
- 
-Selected Intel compiler options: 
- 
-^Option^Description^ 
-|-O0|Disables all optimizations. Recommended for program development and debugging| 
-|-O1|Enables optimization for speed, while being aware of code size (e.g no loop unrolling)| 
-|-O2|Default optimization. Optimizations for speed, including global code scheduling, software pipelining, predication, and speculation.| 
-|**-O3**|-O2 optimizations plus more aggressive optimizations such as prefetching, scalar replacement, and loop transformations. Enables optimizations for technical computing applications (loop-intensive code): loop optimizations and data prefetch.| 
-|**-axAVX**|This option tells the compiler to generate multiple, processor-specific auto-dispatch code paths for Intel processors if there is a performance benefit. It also generates a baseline code path which can run on non-AVX processors. The Intel processor-specific auto-dispatch path is usually more optimized than the baseline path.| 
-|-p |Compiles and links for function profiling with gprof.|  
-|-g| Produces a symbol tables, i.e. line numbers for profiling are available.| 
-|-openmp|Enables the parallelizer to generate multithreaded code based on OpenMP directives.| 
-|-parallel|Tells the auto-parallelizer to generate multithreaded code for loops that can be safely executed in parallel. To use this option, you must also specify -O2 or -O3.| 
-|-opt_report|generate an optimization report to stderr.| 
-|** -fp-model precise**| Allow users to trade off floating point optimizations for accuracy and consistency. Intel recommends specifying ''-fp-model precise''|  
-|**-pc 64**|Force the compiler to use floating double precision (53-bit mantissa) to ensure the portability of results.| 
- 
- 
- 
- 
-Because processors of cluster Lumière are ''Sandybridge''. We strongly recommend the following options to reach the best performances:\\ 
- 
-<WRAP center round box 60%> 
- 
- 
-<color blue>**-axAVX -O3 **</color> 
-</WRAP> 
- 
- 
- 
-if your application is sensitive to the precision, you should add: **-fp-model precise -pc 64**. This option may degrade performances. 
- 
- 
- 
- 
-==== Examples : ==== 
- 
- 
- 
-Compile a C program 
- 
-<code bash> 
-$ icc -axAVX  -O3 -fp-model precise -pc 64 pgm.c -o pgm 
-</code> 
- 
-Compile a Fortran program  
- 
-<code bash> 
-$ ifort -axAVX  -O3 -fp-model precise -pc 64 pgm.f -o pgm 
-</code> 
- 
-Compiler a C++ program 
- 
-<code bash> 
-$ icpc -axAVX  -O3 -fp-model precise -pc 64 pgm.cpp -o pgm 
-</code> 
- 
- 
-Compile an OpenMP program written inc C  
- 
-<code bash> 
-$ icpc -openmp -axAVX -O3 -fp-model precise -pc 64 pgm.cpp -o pgm 
-</code> 
- 
-Compiler an OpenMP program written in Fortran 
- 
-<code bash> 
-$ ifort -openmp -axAVX  O3 -fp-model precise -pc 64 pgm.f -o pgm 
-</code> 
-====== Links ====== 
- 
-  * http://mesowiki.univ-fcomte.fr/intelsuite/get_started_lc.htm 
  • intel-cluster-studio.1398409953.txt.gz
  • Dernière modification : 2015/02/25 08:45
  • (modification externe)