<?xml version="1.0"?>
<feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en">
	<id>http://snipersim.org//w/index.php?action=history&amp;feed=atom&amp;title=Paper%3AIiswc2011Heirman</id>
	<title>Paper:Iiswc2011Heirman - Revision history</title>
	<link rel="self" type="application/atom+xml" href="http://snipersim.org//w/index.php?action=history&amp;feed=atom&amp;title=Paper%3AIiswc2011Heirman"/>
	<link rel="alternate" type="text/html" href="http://snipersim.org//w/index.php?title=Paper:Iiswc2011Heirman&amp;action=history"/>
	<updated>2026-08-12T06:20:46Z</updated>
	<subtitle>Revision history for this page on the wiki</subtitle>
	<generator>MediaWiki 1.34.1</generator>
	<entry>
		<id>http://snipersim.org//w/index.php?title=Paper:Iiswc2011Heirman&amp;diff=211&amp;oldid=prev</id>
		<title>Tcarlson: Created page with &quot;{{DISPLAYTITLE:Using Cycle Stacks to Understand Scaling Bottlenecks in Multi-Threaded Workloads}} Wim Heirman, Trevor E. Carlson, Shuai Che, Kevin Skadron and Lieven Eeckhout  &#039;&#039;...&quot;</title>
		<link rel="alternate" type="text/html" href="http://snipersim.org//w/index.php?title=Paper:Iiswc2011Heirman&amp;diff=211&amp;oldid=prev"/>
		<updated>2012-01-11T14:41:26Z</updated>

		<summary type="html">&lt;p&gt;Created page with &amp;quot;{{DISPLAYTITLE:Using Cycle Stacks to Understand Scaling Bottlenecks in Multi-Threaded Workloads}} Wim Heirman, Trevor E. Carlson, Shuai Che, Kevin Skadron and Lieven Eeckhout  &amp;#039;&amp;#039;...&amp;quot;&lt;/p&gt;
&lt;p&gt;&lt;b&gt;New page&lt;/b&gt;&lt;/p&gt;&lt;div&gt;{{DISPLAYTITLE:Using Cycle Stacks to Understand Scaling Bottlenecks in Multi-Threaded Workloads}}&lt;br /&gt;
Wim Heirman, Trevor E. Carlson, Shuai Che, Kevin Skadron and Lieven Eeckhout&lt;br /&gt;
&lt;br /&gt;
&amp;#039;&amp;#039;Published at the 2011 IEEE International Symposium on Workload Characterization (IISWC)&amp;#039;&amp;#039;&lt;br /&gt;
&lt;br /&gt;
===Abstract===&lt;br /&gt;
This paper proposes a methodology for analyzing parallel performance&lt;br /&gt;
by building cycle stacks. A cycle stack quantifies where the cycles&lt;br /&gt;
have gone, and provides hints towards optimization opportunities.&lt;br /&gt;
We make the case that this is particularly interesting for analyzing&lt;br /&gt;
parallel performance: understanding how cycle components scale with&lt;br /&gt;
increasing core counts and/or input data set sizes leads to insight&lt;br /&gt;
with respect to scaling bottlenecks due to synchronization, load&lt;br /&gt;
imbalance, poor memory performance, etc. &lt;br /&gt;
&lt;br /&gt;
We present several case studies illustrating the use of cycle stacks.&lt;br /&gt;
As a subsequent step, we further extend the methodology to analyze&lt;br /&gt;
sets of parallel workloads using statistical data analysis, and perform&lt;br /&gt;
a workload characterization to understand behavioral differences&lt;br /&gt;
across benchmark suites. We analyze the SPLASH-2, PARSEC and Rodinia&lt;br /&gt;
benchmark suites and conclude that the three benchmark suites cover&lt;br /&gt;
similar areas in the workload space. However, scaling behavior of&lt;br /&gt;
these benchmarks towards larger input sets and/or higher core counts&lt;br /&gt;
is highly dependent on the benchmark, the way in which the inputs&lt;br /&gt;
have been scaled, and on the machine configuration.&lt;br /&gt;
&lt;br /&gt;
===Full text===&lt;br /&gt;
[http://www.exascience.com/wp-content/uploads/2011/09/iiswc2011heirman.pdf Full paper PDF]&lt;br /&gt;
&lt;br /&gt;
===Bibtex entry===&lt;br /&gt;
&amp;lt;pre&amp;gt;@INPROCEEDINGS{heirman2011ucstusbimw,&lt;br /&gt;
  author = {Wim Heirman and Trevor E. Carlson and Shuai Che and Kevin Skadron&lt;br /&gt;
        and Lieven Eeckhout},&lt;br /&gt;
  title = {Using Cycle Stacks to Understand Scaling Bottlenecks in Multi-Threaded&lt;br /&gt;
        Workloads},&lt;br /&gt;
  booktitle = {International Symposium on Workload Characterization (IISWC)},&lt;br /&gt;
  year = {2011},&lt;br /&gt;
  month = nov,&lt;br /&gt;
}&lt;br /&gt;
&amp;lt;/pre&amp;gt;&lt;/div&gt;</summary>
		<author><name>Tcarlson</name></author>
		
	</entry>
</feed>