[llvm-commits] [www-pubs] r98805 - in /www-pubs/trunk: 2010-03-GPGPU-ModelingGPGPU.html 2010-03-GPGPU-ModelingGPGPU.pdf pubs.js

Owen Anderson resistor at mac.com
Wed Mar 17 23:32:13 PDT 2010


Author: resistor
Date: Thu Mar 18 01:32:12 2010
New Revision: 98805

URL: http://llvm.org/viewvc/llvm-project?rev=98805&view=rev
Log:
Add a paper from GPGPU-3.

Added:
    www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.html
    www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.pdf   (with props)
Modified:
    www-pubs/trunk/pubs.js

Added: www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.html
URL: http://llvm.org/viewvc/llvm-project/www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.html?rev=98805&view=auto
==============================================================================
--- www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.html (added)
+++ www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.html Thu Mar 18 01:32:12 2010
@@ -0,0 +1,49 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
+<html>
+<head>
+  <meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
+  <link rel="stylesheet" href="../llvm.css" type="text/css" media="screen">
+  <title>Modeling GPU-CPU Workloads and Systems</title>
+</head>
+<body>
+
+<div class="pub_title">
+  Modeling GPU-CPU Workloads and Systems
+</div>
+<div class="pub_author">
+ Andrew Kerr, Gregory Diamos, and Sudhakar Yalamanchili
+</div>
+
+<h2>Abstract:</h2>
+<blockquote>
+<p>	Heterogeneous systems, systems with multiple processors tailored for specialized tasks, are challenging programming environments. While it may be possible for domain experts to optimize a high performance application for a very specific and well documented system, it may not perform as well or even function on a different system. Developers who have less experience with either the application domain or the system architecture may devote a significant effort to writing a program that merely functions correctly. We believe that a comprehensive analysis and modeling frame-work is necessary to ease application development and automate program optimization on heterogeneous platforms.</p>
+
+<p>This paper reports on an empirical evaluation of 25 CUDA applications on four GPUs and three CPUs, leveraging the Ocelot dynamic compiler infrastructure which can execute and instrument the same CUDA applications on either target. Using a combination of instrumentation and statistical analysis, we record 37 different metrics for each application and use them to derive relationships between program behavior and performance on heterogeneous processors. These relationships are then fed into a modeling frame-work that attempts to predict the performance of similar classes of applications on different processors. Most significantly, this study identifies several non-intuitive relationships between program characteristics and demonstrates that it is possible to accurately model CUDA kernel performance using only metrics that are available before a kernel is executed.</p>
+</blockquote>
+
+<h2>Published:</h2>
+<blockquote>
+  "Modeling GPU-CPU Workloads and Systems"
+  <br>
+  Andrew Kerr, Gregory Diamos, and Sudhakar Yalamanchili
+  <br>
+<i>
+Proc. of the 3rd Workshop on General-Purpose Computation on Graphics Processing Units (GPGPU-3)</i>, Pittsburgh, PA, March 2010.
+</blockquote>
+<h2>Download:</h2>
+<h3>Paper:</h3>
+<ul>
+  <li><a href="2010-03-GPGPU-ModelingGPGPU.pdf">
+  Modeling GPU-CPU Workloads and Systems
+  </a> (PDF)</li>
+</ul>
+
+<!-- *********************************************************************** -->
+<hr>
+  <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
+  src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
+  <a href="http://validator.w3.org/check/referer"><img
+  src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
+
+</body>
+</html>

Added: www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.pdf
URL: http://llvm.org/viewvc/llvm-project/www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.pdf?rev=98805&view=auto
==============================================================================
Binary file - no diff available.

Propchange: www-pubs/trunk/2010-03-GPGPU-ModelingGPGPU.pdf
------------------------------------------------------------------------------
    svn:mime-type = application/octet-stream

Modified: www-pubs/trunk/pubs.js
URL: http://llvm.org/viewvc/llvm-project/www-pubs/trunk/pubs.js?rev=98805&r1=98804&r2=98805&view=diff
==============================================================================
--- www-pubs/trunk/pubs.js (original)
+++ www-pubs/trunk/pubs.js Thu Mar 18 01:32:12 2010
@@ -1,6 +1,13 @@
 // The array should be sorted reverse-chronologically, and will be displayed on
 // the page in the order listed.
 var PUBS = [
+  {url: "2010-03-GPGPU-ModelingGPGPU.html",
+   title: "Modeling GPU-CPU Workloads and Systems",
+   published: "Proc. of the 3rd Workshop on General-Purpose Computation on Graphics Processing Units (GPGPU-3)",
+   author: "Andrew Kerr, Gregory Diamos, and Sudhakar Yalamanchili",
+   location: "Pittsburgh, PA",
+   month: 3,
+   year: 2010},
   {url: "2010-03-ASPLOS-Shoestring.html",
    title: "Shoestring: Probabilistic Soft Error Reliability on the Cheap",
    published: "Proc. of the Fifteenth International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS '10)",





More information about the llvm-commits mailing list