"Fossies" - the Fresh Open Source Software Archive  

Source code changes of the file "README.txt" between
hpcc-1.5.0b.tar.gz and hpcc-1.5.0.tar.gz

About: HPCC (HPC Challenge) benchmark consists of basically 7 tests: HPL, STREAM, RandomAccess, PTRANS, FFTE, DGEMM and b_eff Latency/Bandwidth.

README.txt  (hpcc-1.5.0b):README.txt  (hpcc-1.5.0)
skipping to change at line 207 skipping to change at line 207
- hpcc is the name of the HPC Challenge executable to run. - hpcc is the name of the HPC Challenge executable to run.
After the run, a file called hpccoutf.txt is created. It contains After the run, a file called hpccoutf.txt is created. It contains
results of the benchmark. This file should be uploaded through the web results of the benchmark. This file should be uploaded through the web
form at the HPC Challenge website. form at the HPC Challenge website.
5 Source Code Changes across Versions (ChangeLog) 5 Source Code Changes across Versions (ChangeLog)
*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*= *=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=
5.1 Version 1.5.0beta (2015-07-23) 5.1 Version 1.5.0 (2016-03-18)
===============================
1. Fixed memory leak in STREAM code.
2. Fixed bug in STREAM that resulted in minimum results reported as
0.
3. Removed some of the compilation warnings.
5.2 Version 1.5.0beta (2015-07-23)
=================================== ===================================
1. Added new targets to the main make(1) file. 1. Added new targets to the main make(1) file.
2. Fixed bug introduced while updating to MPI STREAM 1.7 with 2. Fixed bug introduced while updating to MPI STREAM 1.7 with
spurious global communicator (reported by NEC). spurious global communicator (reported by NEC).
3. Added make(1) file for OpenMPI from MacPorts. 3. Added make(1) file for OpenMPI from MacPorts.
4. Fixed bug introduced while updating to MPI STREAM 1.7 that caused 4. Fixed bug introduced while updating to MPI STREAM 1.7 that caused
some ranks to use NULL communicator. some ranks to use NULL communicator.
5. Fixed bug introduced while updating to MPI STREAM 1.7 that caused 5. Fixed bug introduced while updating to MPI STREAM 1.7 that caused
syntax errors. syntax errors.
5.2 Version 1.5.0alpha (2015-05-22) 5.3 Version 1.5.0alpha (2015-05-22)
==================================== ====================================
1. Added global error accounting in STREAM. 1. Added global error accounting in STREAM.
2. Updated checking to report from multiple MPI processes 2. Updated checking to report from multiple MPI processes
contributing to overall error. contributing to overall error.
3. Added barrier to make sure all processes enter STREAM kernel tests 3. Added barrier to make sure all processes enter STREAM kernel tests
at the same time. at the same time.
4. Updated naming conventions to match the original benchmark in 4. Updated naming conventions to match the original benchmark in
STREAM. STREAM.
5. Changed scaling constant to prevent verification from overflowing 5. Changed scaling constant to prevent verification from overflowing
skipping to change at line 241 skipping to change at line 249
6. Simplified MPI communicator code in STREAM. 6. Simplified MPI communicator code in STREAM.
7. Substituted large constants for more descriptive compile time 7. Substituted large constants for more descriptive compile time
arithmetic in STREAM. arithmetic in STREAM.
8. Added the "restrict" keyword to the STREAM vector pointers for 8. Added the "restrict" keyword to the STREAM vector pointers for
faster generated code. faster generated code.
9. Updated STREAM code to the official STREAM MPI version 1.7. 9. Updated STREAM code to the official STREAM MPI version 1.7.
10. Removed infinite loop due to default compiler optimization in 10. Removed infinite loop due to default compiler optimization in
DLAMCH and SLAMCH. DLAMCH and SLAMCH.
11. Added compiler flags to allow compiling with a C++ compiler. 11. Added compiler flags to allow compiling with a C++ compiler.
5.3 Version 1.4.3 (2013-08-26) 5.4 Version 1.4.3 (2013-08-26)
=============================== ===============================
1. Increased the size of scratch vector for local FFT tests that was 1. Increased the size of scratch vector for local FFT tests that was
missed in the previous version (reported by SGI). missed in the previous version (reported by SGI).
2. Added Makefile for Blue Gene/P contributed by Vasil Tsanov. 2. Added Makefile for Blue Gene/P contributed by Vasil Tsanov.
5.4 Version 1.4.2 (2012-10-12) 5.5 Version 1.4.2 (2012-10-12)
=============================== ===============================
1. Increased sizes of scratch vectors for local FFT tests to account 1. Increased sizes of scratch vectors for local FFT tests to account
for runs on systems with large main memory (reported by IBM, SGI and for runs on systems with large main memory (reported by IBM, SGI and
Intel). Intel).
2. Reduced vector size for local FFT tests due to larger scratch 2. Reduced vector size for local FFT tests due to larger scratch
space needed. space needed.
3. Added a type cast to prevent overflow of a 32-bit integer vector 3. Added a type cast to prevent overflow of a 32-bit integer vector
size in FFT data generation routine (reported by IBM). size in FFT data generation routine (reported by IBM).
4. Fixed variable types to handle array sizes that overflow 32-bit 4. Fixed variable types to handle array sizes that overflow 32-bit
integers in RandomAccess (reported by IBM and SGI). integers in RandomAccess (reported by IBM and SGI).
5. Changed time-bound code to be used by default in Global 5. Changed time-bound code to be used by default in Global
RandomAccess and allowed for it to be switched off with a compile RandomAccess and allowed for it to be switched off with a compile
time flag if necessary. time flag if necessary.
6. Code cleanup to allow compilation without warnings of RandomAccess 6. Code cleanup to allow compilation without warnings of RandomAccess
test. test.
7. Changed communication code in PTRANS to avoid large message sizes 7. Changed communication code in PTRANS to avoid large message sizes
that caused problems in some MPI implementations. that caused problems in some MPI implementations.
8. Updated documentation in README.txt and README.html files. 8. Updated documentation in README.txt and README.html files.
5.5 Version 1.4.1 (2010-06-01) 5.6 Version 1.4.1 (2010-06-01)
=============================== ===============================
1. Added optimized variants of RandomAccess that use Linear 1. Added optimized variants of RandomAccess that use Linear
Congruential Generator for random number generation. Congruential Generator for random number generation.
2. Made corrections to comments that provide definition of the 2. Made corrections to comments that provide definition of the
RandomAccess test. RandomAccess test.
3. Removed initialization of the main array from the timed section of 3. Removed initialization of the main array from the timed section of
optimized versions of RandomAccess. optimized versions of RandomAccess.
4. Fixed the length of the vector used to compute error when using 4. Fixed the length of the vector used to compute error when using
MPI implementation from FFTW. MPI implementation from FFTW.
5. Added global reduction to error calculation in MPI FFT to achieve 5. Added global reduction to error calculation in MPI FFT to achieve
more accurate error estimate. more accurate error estimate.
6. Updated documentation in README. 6. Updated documentation in README.
5.6 Version 1.4.0 (2010-03-26) 5.7 Version 1.4.0 (2010-03-26)
=============================== ===============================
1. Added new variant of RandomAccess that uses Linear Congruential 1. Added new variant of RandomAccess that uses Linear Congruential
Generator for random number generation. Generator for random number generation.
2. Rearranged the order of benchmarks so that HPL component runs last 2. Rearranged the order of benchmarks so that HPL component runs last
and may be aborted if the performance of other components was not and may be aborted if the performance of other components was not
satisfactory. RandomAccess is now first to assist in tuning the code. satisfactory. RandomAccess is now first to assist in tuning the code.
3. Added global initialization and finalization routine that allows 3. Added global initialization and finalization routine that allows
to properly initialize and finalize external software and hardware to properly initialize and finalize external software and hardware
components without changing the rest of the HPCC testing harness. components without changing the rest of the HPCC testing harness.
4. Lack of hpccinf.txt is no longer reported as error but as a 4. Lack of hpccinf.txt is no longer reported as error but as a
warning. warning.
5.7 Version 1.3.2 (2009-03-24) 5.8 Version 1.3.2 (2009-03-24)
=============================== ===============================
1. Fixed memory leaks in G-RandomAccess driver routine. 1. Fixed memory leaks in G-RandomAccess driver routine.
2. Made the check for 32-bit vector sizes in G-FFT optional. MKL 2. Made the check for 32-bit vector sizes in G-FFT optional. MKL
allows for 64-bit vector sizes in its FFTW wrapper. allows for 64-bit vector sizes in its FFTW wrapper.
3. Fixed memory bug in single-process FFT. 3. Fixed memory bug in single-process FFT.
4. Update documentation (README). 4. Update documentation (README).
5.8 Version 1.3.1 (2008-12-09) 5.9 Version 1.3.1 (2008-12-09)
=============================== ===============================
1. Fixed a dead-lock problem in FFT component due to use of wrong 1. Fixed a dead-lock problem in FFT component due to use of wrong
communicator. communicator.
2. Fixed the 32-bit random number generator in PTRANS that was using 2. Fixed the 32-bit random number generator in PTRANS that was using
64-bit routines from HPL. 64-bit routines from HPL.
5.9 Version 1.3.0 (2008-11-13) 5.10 Version 1.3.0 (2008-11-13)
=============================== ================================
1. Updated HPL component to use HPL 2.0 source code 1. Updated HPL component to use HPL 2.0 source code
1. Replaced 32-bit Pseudo Random Number Generator (PRNG) with a 1. Replaced 32-bit Pseudo Random Number Generator (PRNG) with a
64-bit one. 64-bit one.
2. Removed 3 numerical checks of the solution residual with a 2. Removed 3 numerical checks of the solution residual with a
single one. single one.
3. Added support for 64-bit systems with large memory sizes 3. Added support for 64-bit systems with large memory sizes
(before they would overflow during index calculations 32-bit (before they would overflow during index calculations 32-bit
integers.) integers.)
2. Introduced a limit on FFT vector size so they fit in a 32-bit 2. Introduced a limit on FFT vector size so they fit in a 32-bit
integer (only applicable when using FFTW version 2.) integer (only applicable when using FFTW version 2.)
5.10 Version 1.2.0 (2007-06-25) 5.11 Version 1.2.0 (2007-06-25)
================================ ================================
1. Changes in the FFT component: 1. Changes in the FFT component:
1. Added flexibility in choosing vector sizes and processor 1. Added flexibility in choosing vector sizes and processor
counts: now the code can do powers of 2, 3, and 5 both counts: now the code can do powers of 2, 3, and 5 both
sequentially and in parallel tests. sequentially and in parallel tests.
2. FFTW can now run with ESTIMATE (not just MEASURE) flag: it 2. FFTW can now run with ESTIMATE (not just MEASURE) flag: it
might produce worse performance results but often reduces time to might produce worse performance results but often reduces time to
run the test and cuases less memory fragmentation. run the test and cuases less memory fragmentation.
skipping to change at line 383 skipping to change at line 391
6. Miscellaneous changes: 6. Miscellaneous changes:
1. Added better support for Windows-based clusters by taking 1. Added better support for Windows-based clusters by taking
advantage of Win32 API. advantage of Win32 API.
2. Added custom memory allocator to deal with memory fragmentation 2. Added custom memory allocator to deal with memory fragmentation
on some systems. on some systems.
3. Added better reporting of configuration options in the output 3. Added better reporting of configuration options in the output
file. file.
5.11 Version 1.0.0 (2005-06-11) 5.12 Version 1.0.0 (2005-06-11)
================================ ================================
5.12 Version 0.8beta (2004-10-19) 5.13 Version 0.8beta (2004-10-19)
================================== ==================================
5.13 Version 0.8alpha (2004-10-15) 5.14 Version 0.8alpha (2004-10-15)
=================================== ===================================
5.14 Version 0.6beta (2004-08-21) 5.15 Version 0.6beta (2004-08-21)
================================== ==================================
5.15 Version 0.6alpha (2004-05-31) 5.16 Version 0.6alpha (2004-05-31)
=================================== ===================================
5.16 Version 0.5beta (2003-12-01) 5.17 Version 0.5beta (2003-12-01)
================================== ==================================
5.17 Version 0.4alpha (2003-11-13) 5.18 Version 0.4alpha (2003-11-13)
=================================== ===================================
5.18 Version 0.3alpha (2004-11-05) 5.19 Version 0.3alpha (2004-11-05)
=================================== ===================================
----------------------------------------------------------------------- -----------------------------------------------------------------------
This document was translated from LaTeX by HeVeA (2). This document was translated from LaTeX by HeVeA (2).
----------------------------------- -----------------------------------
(1) University of Tennessee Knoxville, Innovative Computing Laboratory (1) University of Tennessee Knoxville, Innovative Computing Laboratory
(2) http://hevea.inria.fr/index.html (2) http://hevea.inria.fr/index.html
 End of changes. 19 change blocks. 
20 lines changed or deleted 27 lines changed or added

Home  |  About  |  All  |  Newest  |  Fossies Dox  |  Screenshots  |  Comments  |  Imprint  |  Privacy  |  HTTPS