Différences

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

Lien vers cette vue comparative

Prochaine révision
Révision précédente
compilation:intel-cluster-studio [2015/02/27 13:17] – créée admincompilation:intel-cluster-studio [2017/05/05 10:43] (Version actuelle) – [Compilers Options] admin
Ligne 1: Ligne 1:
-====== Intel Cluster Studio XE 2013 ======+====== Intel Cluster Studio XE 2013 (upgrade 15.0.3) ======
  
 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 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:
Ligne 20: Ligne 20:
 Before you start compiling your application, you need to load the corresponding module: Before you start compiling your application, you need to load the corresponding module:
  
-<code>+<code bash>
  
-module load intel/14.0.2+$ module load intel/14.0.2
  
 </code>  </code> 
Ligne 48: Ligne 48:
 |-fast|Shorthand. ''-O3 -ipo -static -xHOST -no-prec-div''. Note that the processor specific optimization flag (''-xHOST'') will optimize for the processor compiled on it is the only flag of ''-fast'' that may be overridden.| |-fast|Shorthand. ''-O3 -ipo -static -xHOST -no-prec-div''. Note that the processor specific optimization flag (''-xHOST'') will optimize for the processor compiled on it is the only flag of ''-fast'' that may be overridden.|
 |**-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.| |**-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.|+|**-axCORE-AVX2**|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.|  |-p |Compiles and links for function profiling with gprof.| 
 |-g| Produces a symbol tables, i.e. line numbers for profiling are available.| |-g| Produces a symbol tables, i.e. line numbers for profiling are available.|
Ligne 65: Ligne 65:
  
  
-<color blue>**-axAVX -O3 **</color>+<color blue>**-axCORE-AVX2 -O2 **</color>
 </WRAP> </WRAP>
  
Ligne 82: Ligne 82:
  
 <code bash> <code bash>
-$ icc -axAVX  -O3 -fp-model precise -pc 64 pgm.c -o pgm+$ icc -axCORE-AVX2  -O3 -fp-model precise -pc 64 pgm.c -o pgm
 </code> </code>
  
Ligne 88: Ligne 88:
  
 <code bash> <code bash>
-$ ifort -axAVX  -O3 -fp-model precise -pc 64 pgm.f -o pgm+$ ifort -axCORE-AVX2  -O3 -fp-model precise -pc 64 pgm.f -o pgm
 </code> </code>
  
Ligne 94: Ligne 94:
  
 <code bash> <code bash>
-$ icpc -axAVX  -O3 -fp-model precise -pc 64 pgm.cpp -o pgm+$ icpc -axCORE-AVX2  -O3 -fp-model precise -pc 64 pgm.cpp -o pgm
 </code> </code>
  
Ligne 101: Ligne 101:
  
 <code bash> <code bash>
-$ icpc -openmp -axAVX -O3 -fp-model precise -pc 64 pgm.cpp -o pgm+$ icpc -openmp -axCORE-AVX2 -O3 -fp-model precise -pc 64 pgm.cpp -o pgm
 </code> </code>
  
Ligne 107: Ligne 107:
  
 <code bash> <code bash>
-$ ifort -openmp -axAVX  -O3 -fp-model precise -pc 64 pgm.f -o pgm+$ ifort -openmp -axCORE-AVX2  -O3 -fp-model precise -pc 64 pgm.f -o pgm
 </code> </code>
 ====== Compiler Auto Vectorization ====== ====== Compiler Auto Vectorization ======
Ligne 114: Ligne 114:
 Using the ''-xHost'' flag enables the highest level of vectorization supported on the processor on which the user compiles. Note that the Intel compiler will try to vectorize a code with SSE2 instructions at optimizations of ''-O2'' or higher. Disable this by specifying ''-no-vec''.  Using the ''-xHost'' flag enables the highest level of vectorization supported on the processor on which the user compiles. Note that the Intel compiler will try to vectorize a code with SSE2 instructions at optimizations of ''-O2'' or higher. Disable this by specifying ''-no-vec''. 
  
-The Intel compiler can generate a single executable with multiple levels of vectorization with the ''-ax'' flag, which takes the same options as the ''-x'' flag (i.e., AVX, ..., SSE2). This flag will generate run-time checks to determine the level of vectorization support on the processor and  ill then choose the optimal execution path for that processor. It will also generate a baseline execution path that is taken if the ''-ax'' level of vectorization specified is not supported. Example ''-axAVX''. +The Intel compiler can generate a single executable with multiple levels of vectorization with the ''-ax'' flag, which takes the same options as the ''-x'' flag (i.e.,AVX2 ,AVX, ..., SSE2). This flag will generate run-time checks to determine the level of vectorization support on the processor and  ill then choose the optimal execution path for that processor. It will also generate a baseline execution path that is taken if the ''-ax'' level of vectorization specified is not supported. Example ''-axCORE-AVX2''. 
  
 Another useful option for the Intel compiler is the ''-vec-report'' flag, which generates diagnostic information regarding vectorization to ''stdout''. The ''-vec-report'' flag takes an optional parameter that can be a number between 0 and 5 (e.g., ''-vec-report0''), with 0 disabling diagnostics and 5 providing the most detailed diagnostics about what loops were optimized, what loops were not optimized, and why those loops were not optimized. The output can be useful to identify possible strategies to get a loop to vectorize. Another useful option for the Intel compiler is the ''-vec-report'' flag, which generates diagnostic information regarding vectorization to ''stdout''. The ''-vec-report'' flag takes an optional parameter that can be a number between 0 and 5 (e.g., ''-vec-report0''), with 0 disabling diagnostics and 5 providing the most detailed diagnostics about what loops were optimized, what loops were not optimized, and why those loops were not optimized. The output can be useful to identify possible strategies to get a loop to vectorize.
  
 +===== Example =====
 +Comparing a program, Matrix Multiply (sequential version), compiled with with options: ''-O3'' and ''-03 -xCORE-AVX2'' for **N=6000**
  
 +^Option^Time (s) ^ MFlops ^
 +| ''-O3'' | 145 | 4079 |
 +|''-O3 -xCORE-AVX2''| 49 |4405 |
 ====== Links ====== ====== Links ======
  
   *https://software.intel.com/en-us/intel-sdp-home   *https://software.intel.com/en-us/intel-sdp-home
   * http://mesowiki.univ-fcomte.fr/intelsuite/get_started_lc.htm   * http://mesowiki.univ-fcomte.fr/intelsuite/get_started_lc.htm
  • compilation/intel-cluster-studio.1425043063.txt.gz
  • Dernière modification : 2015/02/27 13:17
  • de admin