Gecko [ v99win.today ]

Name	Size	Permission
images	[ DIR ]	drwxr-xr-x
wand	[ DIR ]	drwxr-xr-x
ChangeLog-2001.html	23.56 KB	-rw-r--r--
ChangeLog-2002.html	95.11 KB	-rw-r--r--
ChangeLog-2003.html	257.78 KB	-rw-r--r--
ChangeLog-2004.html	87.64 KB	-rw-r--r--
ChangeLog-2005.html	58.03 KB	-rw-r--r--
ChangeLog-2006.html	14.43 KB	-rw-r--r--
ChangeLog-2007.html	57.83 KB	-rw-r--r--
ChangeLog-2008.html	127.75 KB	-rw-r--r--
ChangeLog-2009.html	106.29 KB	-rw-r--r--
ChangeLog-2010.html	44.81 KB	-rw-r--r--
ChangeLog-2011.html	40.59 KB	-rw-r--r--
ChangeLog-2012.html	58.38 KB	-rw-r--r--
ChangeLog-2013.html	35.93 KB	-rw-r--r--
ChangeLog-2014.html	57.66 KB	-rw-r--r--
ChangeLog-2015.html	111.57 KB	-rw-r--r--
ChangeLog-2016.html	59.17 KB	-rw-r--r--
ChangeLog-2017.html	72.3 KB	-rw-r--r--
ChangeLog-2018.html	170.85 KB	-rw-r--r--
ChangeLog-2019.html	84.87 KB	-rw-r--r--
ChangeLog-2020.html	74.61 KB	-rw-r--r--
ChangeLog-2021.html	50.82 KB	-rw-r--r--
Changes.html	4.78 KB	-rw-r--r--
Copyright.html	15.27 KB	-rw-r--r--
FAQ.html	47.41 KB	-rw-r--r--
GraphicsMagick.html	376.79 KB	-rw-r--r--
Hg.html	15.76 KB	-rw-r--r--
INSTALL-unix.html	60.78 KB	-rw-r--r--
INSTALL-windows.html	40.33 KB	-rw-r--r--
ImageMagickObject.html	7.17 KB	-rw-r--r--
NEWS.html	211.05 KB	-rw-r--r--
OpenMP.html	12.77 KB	-rw-r--r--
README.html	22.68 KB	-rw-r--r--
animate.html	41.85 KB	-rw-r--r--
authors.html	8.6 KB	-rw-r--r--
batch.html	10.6 KB	-rw-r--r--
benchmark.html	11.78 KB	-rw-r--r--
benchmarks.html	7.1 KB	-rw-r--r--
bugs.html	2.89 KB	-rw-r--r--
color.html	30.19 KB	-rw-r--r--
compare.html	19.27 KB	-rw-r--r--
composite.html	39.9 KB	-rw-r--r--
configure-target-setup.png	7.21 KB	-rw-r--r--
conjure.html	15.4 KB	-rw-r--r--
contribute.html	6.85 KB	-rw-r--r--
convert.html	90.16 KB	-rw-r--r--
display.html	140.58 KB	-rw-r--r--
docutils-api.css	8.56 KB	-rw-r--r--
docutils-articles.css	9.95 KB	-rw-r--r--
download.html	7.25 KB	-rw-r--r--
favicon.ico	922 B	-rw-r--r--
formats.html	41.29 KB	-rw-r--r--
gm.html	891.41 KB	-rw-r--r--
identify.html	13.4 KB	-rw-r--r--
import.html	30.34 KB	-rw-r--r--
index.html	9.75 KB	-rw-r--r--
links.html	14.44 KB	-rw-r--r--
magick.css	2.36 KB	-rw-r--r--
miff.html	17.44 KB	-rw-r--r--
mission.html	3.32 KB	-rw-r--r--
mogrify.html	81.71 KB	-rw-r--r--
montage.html	54.05 KB	-rw-r--r--
motion-picture.html	32.02 KB	-rw-r--r--
perl.html	54.33 KB	-rw-r--r--
process.html	5.97 KB	-rw-r--r--
programming.html	5.74 KB	-rw-r--r--
project.html	3.82 KB	-rw-r--r--
quantize.html	11.81 KB	-rw-r--r--
reference.html	2.42 KB	-rw-r--r--
security.html	19.6 KB	-rw-r--r--
smile.c.gz	982 B	-rw-r--r--
thanks.html	6.07 KB	-rw-r--r--
time.html	4.77 KB	-rw-r--r--
tools.html	11.6 KB	-rw-r--r--
utilities.html	4.99 KB	-rw-r--r--
version.html	5.98 KB	-rw-r--r--

Code Editor : OpenMP.html

<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="generator" content="Docutils 0.16: http://docutils.sourceforge.net/" />
<title>OpenMP in GraphicsMagick</title>
<link rel="stylesheet" href="docutils-articles.css" type="text/css" />
</head>
<body>

<div class="banner">
<img src="images/gm-107x76.png" alt="GraphicMagick logo" width="107" height="76" />
<span class="title">GraphicsMagick</span>
<form action="http://www.google.com/search">
  <input type="hidden" name="domains" value="www.graphicsmagick.org" />
  <input type="hidden" name="sitesearch" value="www.graphicsmagick.org" />
<span class="nowrap"><input type="text" name="q" size="25" maxlength="255" />&nbsp;<input type="submit" name="sa" value="Search" /></span>
</form>
</div>

<div class="navmenu">
<ul>
  <li><a href="index.html">Home</a></li>
  <li><a href="project.html">Project</a></li>
  <li><a href="download.html">Download</a></li>
  <li><a href="README.html">Install</a></li>
  <li><a href="Hg.html">Source</a></li>
  <li><a href="NEWS.html">News</a> </li>
  <li><a href="utilities.html">Utilities</a></li>
  <li><a href="programming.html">Programming</a></li>
  <li><a href="reference.html">Reference</a></li>
</ul>
</div>
<div class="document" id="openmp-in-graphicsmagick">
<h1 class="title">OpenMP in GraphicsMagick</h1>

<div class="contents local topic" id="contents">
<ul class="simple">
<li><a class="reference internal" href="#overview" id="id1">Overview</a></li>
<li><a class="reference internal" href="#limitations" id="id2">Limitations</a></li>
<li><a class="reference internal" href="#openmp-variables" id="id3">OpenMP Variables</a></li>
</ul>
</div>
<div class="section" id="overview">
<h1><a class="toc-backref" href="#id1">Overview</a></h1>
<p>GraphicsMagick has been transformed to use <a class="reference external" href="http://openmp.org/">OpenMP</a> for the 1.3 release
series. OpenMP is a portable framework for accelerating CPU-bound and
memory-bound operations using multiple threads. OpenMP originates in
the super-computing world and has been available in one form or
another since the late '90s.</p>
<p>Since GCC 4.2 has introduced excellent OpenMP support via <a class="reference external" href="http://gcc.gnu.org/onlinedocs/libgomp/">GOMP</a>,
OpenMP has become available to the masses.  Recently, <a class="reference external" href="https://clang.llvm.org/">Clang</a> has
also implemented good OpenMP support. Microsoft Visual Studio
Professional 2005 and later support OpenMP so Windows users can
benefit as well. Any multi-CPU and/or multi-core system is potentially
a good candidate for use with OpenMP.  Modern multi-core chipsets from
AMD, Intel, IBM, Oracle, and ARM perform very well with OpenMP.</p>
<p>Most image processing routines are comprised of loops which iterate
through the image pixels, image rows, or image regions. These loops
are accelerated using OpenMP by executing portions of the total loops
in different threads, and therefore on a different processor
core. CPU-bound algorithms benefit most from OpenMP, but memory-bound
algorithms may also benefit as well since the memory is accessed by
different CPU cores, and sometimes the CPUs have their own path to
memory. For example, the AMD Opteron is a <a class="reference external" href="https://en.wikipedia.org/wiki/Non-uniform_memory_access">NUMA</a> (Non-Uniform Memory
Architecture) design such that multi-CPU systems split the system
memory across CPUs so each CPU adds more memory bandwidth as well.
Server-class CPUs offer more independent memory channels than desktop
CPUs do.</p>
<p>For severely CPU-bound algorithms, it is not uncommon to see a linear
speed-up (within the constraints of <a class="reference external" href="https://en.wikipedia.org/wiki/Amdahl%27s_law">Amdahl's law</a>) due to the number
of cores. For example, a two core system executes the algorithm twice
as fast, and a four core system executes the algorithm four times as
fast. Memory-bound algorithms scale based on the memory bandwith
available to the cores. For example, memory-bound algorithms scale up
to almost 1.5X on my four core Opteron system due to its <a class="reference external" href="https://en.wikipedia.org/wiki/Non-uniform_memory_access">NUMA</a>
architecture. Some systems/CPUs are able to immediately context switch
to another thread if the core would be blocked waiting for memory,
allowing multiple memory accesses to be pending at once, and thereby
improving throughput.  For example, typical speedup of 20-32X (average
24X) has been observed on the Sun SPARC T2 CPU, which provides 8
cores, with 8 virtual CPUs per core (64 threads).</p>
<p>An approach used in GraphicsMagick is to recognize the various access
patterns in the existing code, and re-write the algorithms (sometimes
from scratch) to be based on a framework that we call &quot;pixel iterators&quot;.
With this approach, the computation is restricted to a small unit (a
callback function) with very well defined properties, and no knowledge as
to how it is executed or where the data comes from. This approach removes
the loops from the code and puts the loops in the framework, which may be
adjusted based on experience. The continuing strategy will be to
recognize design patterns and build frameworks which support those
patterns. Sometimes algorithms are special/exotic enough that it is much
easier to instrument the code for OpenMP rather than to attempt to fit
the algorithm into a framework.</p>
<p>Since OpenMP is based on multi-threading, multiple threads access the
underlying pixel storage at once. The interface to this underlying
storage is called the &quot;pixel cache&quot;. The original pixel cache code
(derived from ImageMagick) was thread safe only to the extent that it
allowed one thread per image. This code has now been re-written so that
multiple threads may safely and efficiently work on the pixels in one
image. The re-write also makes the pixel cache thread safe if a
multi-threaded application uses an OpenMP-fortified library.</p>
<p>GraphicsMagick provides its own built-in 'benchmark' driver utility
which may be used to execute a multi-threaded benchmark of any other
utility command.</p>
<p>Using the built-in 'benchmark' driver utility, the following is an
example of per-core speed-up due to OpenMP on a four-core AMD Opteron
system (with Firefox and other desktop software still running).  The
image is generated dynamically based on the 'granite' pattern and all
the pixel quantum values have 30% gaussian noise added:</p>
<pre class="literal-block">
% gm benchmark -stepthreads 1 -duration 10 convert \
  -size 2048x1080 pattern:granite -operator all Noise-Gaussian 30% null:
Results: 1 threads 5 iter 11.34s user 11.340000s total 0.441 iter/s 0.441 iter/cpu 1.00 speedup 1.000 karp-flatt
Results: 2 threads 9 iter 20.34s user 10.190000s total 0.883 iter/s 0.442 iter/cpu 2.00 speedup 0.000 karp-flatt
Results: 3 threads 14 iter 31.72s user 10.600000s total 1.321 iter/s 0.441 iter/cpu 3.00 speedup 0.001 karp-flatt
Results: 4 threads 18 iter 40.84s user 10.460000s total 1.721 iter/s 0.441 iter/cpu 3.90 speedup 0.008 karp-flatt
</pre>
<p>Note that the &quot;iter/s cpu&quot; value is a measure of the number of
iterations given the amount of reported CPU time consumed. It is an
effective measure of relative efficacy since its value should ideally
not drop as iterations are added.  The <a class="reference external" href="https://en.wikipedia.org/wiki/Karp%E2%80%93Flatt_metric">karp-flatt metric</a> is another
useful metric for evaluating thread-speedup efficiency. In the above
example, the total speedup was about 3.9X with only a slight loss of
CPU efficiency as threads are added.</p>
</div>
<div class="section" id="limitations">
<h1><a class="toc-backref" href="#id2">Limitations</a></h1>
<p>Often it is noticed that the memory allocation functions (e.g. from
the standard C library such as GNU libc) significantly hinder
performance since they are designed or optimized for single-threaded
programs, or prioritize returning memory to the system over speed.
Memory allocators are usually designed and optimized for programs
which perform thousands of small allocations, and if they make a large
memory allocation, they retain that memory for a long time.
GraphicsMagick performs large memory allocations for raster image
storage interspersed with a limited number of smaller allocations for
supportive data structures.  This memory is released very quickly
since GraphicsMagick is highly optimized and thus the time between
allocation and deallocation can be very short.  It has been observed
that some memory allocators are much slower to allocate and deallocate
large amounts of memory (e.g. a hundred megabytes) than alternative
allocators, even in single-threaded programs.  Under these conditions,
the program can spend considerable time mysteriously &quot;sleeping&quot;.</p>
<p>In order to help surmount problems with the default memory allocators,
the configure script offers support for use of Google <a class="reference external" href="https://github.com/gperftools/gperftools">gperftools</a> <a class="reference external" href="https://github.com/gperftools/gperftools/wiki">'tcmalloc'</a>, Solaris mtmalloc,
and Solaris umem libraries via the --with-tcmalloc, --with-mtmalloc,
and --with-umem options, respectively.  When the allocation functions
are behaving badly, the memory allocation/deallocation performance
does not scale as threads are added and thus additional threads spend
more time sleeping (e.g. on a lock, or in munmap()) rather than doing
more work.  Performance improvements of a factor of two are not
uncommon even before contending with the hugh CPU core/thread counts
available on modern CPUs.  Using more threads which are slowed by
poorly-matched memory allocation functions is wasteful of memory,
system resources, human patience, and electrical power.</p>
<p>Many modern CPUs support &quot;Turbo&quot; modes where the CPU clock rate is
boosted if only a few cores are active.  When a CPU provides a &quot;Turbo&quot;
mode, this decreases the apparent speed-up compared to using one
thread because the one thread was executed at a much higher clock
rate.  Likewise, when a CPU becomes very hot (due to being heavily
used), it may decrease its clock rates overall to avoid burning up,
and this may also decreases the actual speed-up when using many
threads compared to using one thread.  Many CPUs support
&quot;hyperthreads&quot; or other mechanisms in which one physical core will
support multiple light-weight threads, and if the core is efficiently
used by one thread, then this will decrease the apparent per-thread
speed-up but the peak speed-up will hopefully still be bounded by the
number of physical cores.</p>
<p>In most cases, OpenMP does not speed-up loading an image from a file,
or writing an image to a file.  It is common for file decode and
encode to take longer than processing the image.  Using uncompressed
formats is recommended with a fast I/O subsystem (or in-memory 'blobs'
in order to obtain the greated speed-up from OpenMP.</p>
<p>It has been observed that sometimes it takes much longer to start and
stop GraphicsMagick than it takes for it to run the requested
algorithm.  The slowness is due to inefficiencies of the libraries
that GraphicsMagick is linked with (especially the ICU library that
libxml2 is often linked with).  If GraphicsMagick takes too long to
perform trivial operations, then consider using the 'modules' build,
and investigate the 'batch' utility which allows running many
GraphicsMagick commands as a 'batch' script.  If a 'modules' build is
not feasible, then configuring GraphicsMagick to only support the
specific formats actually needed can help with its execution time and
improve opportunity for OpenMP speed-up.</p>
</div>
<div class="section" id="openmp-variables">
<h1><a class="toc-backref" href="#id3">OpenMP Variables</a></h1>
<p>According to the OpenMP specification, the OMP_NUM_THREADS evironment
variable may be used to specify the number of threads available to the
application. Typically this is set to the number of processor cores on
the system but may be set lower to limit resource consumption or (in
some cases) to improve execution efficiency.  The GraphicsMagick
commands also accept a <tt class="docutils literal"><span class="pre">-limit</span> threads limit</tt> type option for
specifying the maximum number of threads to use.</p>
</div>
</div>

${this.title}

Code Editor : OpenMP.html