TCE-IP

From HPCRL Wiki
(Difference between revisions)
Jump to: navigation, search
(ogGMlmic)
 
(926 intermediate revisions by 4 users not shown)
Line 1: Line 1:
http://okis6027.phreesite.com/ coloured contact lenses
+
=== Introduction ===
http://mike9045.110mb.com/ free video file converter
+
The IP-TCE library is a set of C functions designed to provide
http://kiva3512.freehostia.com/ buy mercenaries 2  
+
high-performance array index permutation (index sorting), an important
http://maso0237.mycyberway.com/ xbox 360 gaming console
+
group of kernels used in many scientific applications and compilation
http://buss9103.fizwig.com/ windows xp and download
+
techniques. The generation of the library code combines analytical and
http://polb3502.freehostguy.com/ jessica simpson wedding hair
+
empirical approaches. The details of how to generate the efficient 2-D IP
http://nemo9045.007gb.com/ www youtube downloader com
+
(Matrix Transposition) code can be found in [1]. For high-dimensional arrays,
http://okis6027.phreesite.com/plano-colored-contact-lenses.html colored contact lenses uk
+
with challenges such as high indexing costs and short dimensions, we employ
http://mike9045.110mb.com/free-video-converter-for-vista.html free video converter for vista
+
optimizations such as restricting code versions, using one-level tiling and
http://kiva3512.freehostia.com/mercenaries-2-reviews.html mercenaries 2 world in flames videos
+
generating indexing code to achieve high performance without code size
http://maso0237.mycyberway.com/which-xbox-360-console.html xbox 360 console release date
+
explosion.
http://buss9103.fizwig.com/download-service-pack-2-for-windows-xp.html windows xp sp 1 download
+
 
http://polb3502.freehostguy.com/jessica-simpson-hair-colors.html jessica simpson hair exstentions
+
The source code we provide does index permutation on 2-D,4-D and 6-D
http://nemo9045.007gb.com/psp-youtube-downloader.html use youtube downloader
+
64-bit floating point arrays on IA-32 machines. Except for the 2-D functions that
http://okis6027.phreesite.com/multifocal-colored-contact-lenses.html trial color contact lenses
+
have more complicated optimizations than the others, the library employs SSE2
http://mike9045.110mb.com/free-video-converter-reviews.html a good free video converter
+
instructions when the fastest varying dimensions of source and destination
http://kiva3512.freehostia.com/2-mercenaries-on.html for mercenaries 2 world in flames
+
arrays are multiples of vector sizes in elements, otherwise a scalar version
http://maso0237.mycyberway.com/xbox-360-console-crate.html xbox 360 console pics
+
is chosen. The user should be able to modify our code to get IP code working for
http://buss9103.fizwig.com/buy-and-download-windows-xp.html windows xp service pack 2 full download
+
other dimension numbers.
http://polb3502.freehostguy.com/are-jessica-simpson-hair.html jessica simpson hair extenions
+
 
http://nemo9045.007gb.com/youtube-downloader-work.html youtube downloader by
+
The IP-TCE library implements two variants of index permutation. One variant
http://okis6027.phreesite.com/wild-color-contact-lenses.html free acuvue contact lenses
+
implements B = factor * Permute(A, permutation) and the other one is the
http://mike9045.110mb.com/download-total-video-converter.html a video converter
+
accumulative version B = B + factor * Permute(A, permutation). Doing so makes
http://kiva3512.freehostia.com/in-mercenaries-2-world-in-flames.html mercenaries 2 at
+
our code compatible with the index permutation routines in nwchem/tce. We have
http://maso0237.mycyberway.com/console-number-on-xbox-360.html xbox 360 console to console
+
been able to plug our code into nwchem/tce and obtained overall performance
http://buss9103.fizwig.com/download-of-windows-xp-service-pack-2.html download windows xp pro service pack 3
+
improvements ranging from 74% to 253% with different methods and inputs.
http://polb3502.freehostguy.com/carry-jessica-simpson-hair.html i buy jessica simpson hair
+
 
http://nemo9045.007gb.com/tech-crunch-youtube-downloader.html youtube downloader unable
+
=== Compilation ===
http://okis6027.phreesite.com/acuvue-soft-contact-lenses.html acuvue advanced contact lenses
+
The library code can be compiled by either the Intel C compiler or the GNU C
http://mike9045.110mb.com/best-video-converter.html video converter software
+
compiler. We used icc 10.1 and gcc 4.1.2 when testing the code. If other
http://kiva3512.freehostia.com/mercenaries-2-for-the.html run mechwarrior 2 mercenaries
+
compilers such as pgcc is used, the user is responsible for finding the
http://maso0237.mycyberway.com/mass-effect-branded-xbox-360-console.html cheats for turok on xbox 360
+
alignment directives used by the compiler and changing the library code
http://buss9103.fizwig.com/free-download-windows-xp-upgrade.html windows media center xp free download
+
correspondingly.
http://polb3502.freehostguy.com/jessica-alba-hair-how.html jessica alba hair pics
+
 
http://nemo9045.007gb.com/youtube-google-video-downloader-for-zune.html youtube downloader video url
+
=== Library Usage ===
http://mike9045.110mb.com/mediacell-video-converter.html video converter plus
+
The interface of the IP-TCE library is compatible with the index permutation
http://mike9045.110mb.com/video-converter-torrents.html pavtube video converter
+
routines in nwchem/tce and is mainly for Fortran used. For example,
 +
the prototype of the 4-D non-accumulative permutation routine is:
 +
 
 +
                  tce_sort_4_(double* unsorted,double* sorted,
 +
                              int* a_in, int* b_in, int* c_in, int* d_in,
 +
                              int* i_in, int* j_in, int* k_in, int* l_in,
 +
                              double* factor_in)
 +
 
 +
Where all the arguments are pointers and function names are in the lowercase
 +
and end with "_". If needed,  the user can write a wrapper or directly modify
 +
the function (which is actually a wrapper to the SIMD and scalar IP functions)
 +
to obtain a desired interface.
 +
 
 +
=== Contact Info ===
 +
Please contact Qingda Lu(luq@cse.ohio-state.edu) for questions.
 +
 
 +
=== Reference ===
 +
[1] Qingda Lu, Sriram Krishnamoorthy, P. Sadayappan: Combining analytical and
 +
empirical approaches in tuning matrix transposition. 15th International
 +
Conference on Parallel Architecture and Compilation Techniques(PACT 2006):233-242

Latest revision as of 23:34, 20 February 2009

Contents

Introduction

The IP-TCE library is a set of C functions designed to provide high-performance array index permutation (index sorting), an important group of kernels used in many scientific applications and compilation techniques. The generation of the library code combines analytical and empirical approaches. The details of how to generate the efficient 2-D IP (Matrix Transposition) code can be found in [1]. For high-dimensional arrays, with challenges such as high indexing costs and short dimensions, we employ optimizations such as restricting code versions, using one-level tiling and generating indexing code to achieve high performance without code size explosion.

The source code we provide does index permutation on 2-D,4-D and 6-D 64-bit floating point arrays on IA-32 machines. Except for the 2-D functions that have more complicated optimizations than the others, the library employs SSE2 instructions when the fastest varying dimensions of source and destination arrays are multiples of vector sizes in elements, otherwise a scalar version is chosen. The user should be able to modify our code to get IP code working for other dimension numbers.

The IP-TCE library implements two variants of index permutation. One variant implements B = factor * Permute(A, permutation) and the other one is the accumulative version B = B + factor * Permute(A, permutation). Doing so makes our code compatible with the index permutation routines in nwchem/tce. We have been able to plug our code into nwchem/tce and obtained overall performance improvements ranging from 74% to 253% with different methods and inputs.

Compilation

The library code can be compiled by either the Intel C compiler or the GNU C compiler. We used icc 10.1 and gcc 4.1.2 when testing the code. If other compilers such as pgcc is used, the user is responsible for finding the alignment directives used by the compiler and changing the library code correspondingly.

Library Usage

The interface of the IP-TCE library is compatible with the index permutation routines in nwchem/tce and is mainly for Fortran used. For example, the prototype of the 4-D non-accumulative permutation routine is:

                  tce_sort_4_(double* unsorted,double* sorted,
                              int* a_in, int* b_in, int* c_in, int* d_in,
                              int* i_in, int* j_in, int* k_in, int* l_in,
                              double* factor_in)

Where all the arguments are pointers and function names are in the lowercase and end with "_". If needed, the user can write a wrapper or directly modify the function (which is actually a wrapper to the SIMD and scalar IP functions) to obtain a desired interface.

Contact Info

Please contact Qingda Lu(luq@cse.ohio-state.edu) for questions.

Reference

[1] Qingda Lu, Sriram Krishnamoorthy, P. Sadayappan: Combining analytical and empirical approaches in tuning matrix transposition. 15th International Conference on Parallel Architecture and Compilation Techniques(PACT 2006):233-242