Hi all,<br><br>
> > Are the scripts used to do that benchmark available? (I know<br>> > XAutomation is already available in Debian)<br>
><br>> I am not much aware of testing tools they used. Maybe you can contact<br>
other authors of this paper for details.<br><br>Just a comment, for this paper we used Nokia N800(ARM) as target. <br>We did not have an automated script to benchmark used. Please, take a look<br>on paper and let me know if you have any problem to reproduce it.<br>
<br>-- Allan Bezerra<br>INdT-Nokia-Brazil<br><br><div class="gmail_quote">On Thu, Apr 10, 2008 at 1:08 PM, Nitin Gupta <<a href="mailto:nitingupta910@gmail.com">nitingupta910@gmail.com</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="Ih2E3d">On Thu, Apr 10, 2008 at 12:28 PM, John McCabe-Dansted <<a href="mailto:gmatht@gmail.com">gmatht@gmail.com</a>> wrote:<br>
> On Thu, Apr 10, 2008 at 12:04 AM, Nitin Gupta <<a href="mailto:nitingupta910@gmail.com">nitingupta910@gmail.com</a>> wrote:<br>
> > So, I will require performance numbers to justify inclusion of compcache<br>
> > in mainline. We already have separate perf numbers for lzo(compressor) and<br>
> > tlsf(allocator) but none that shows that using compcache is good for<br>
> > perf under memory pressure.<br>
><br>
> Well, the my testing showing that compcache is more robust with that<br>
> with no swap at all under low memory situations. This would have been<br>
> enough for Ubuntu if not for the feature freeze. In any case,<br>
> benchmarks would be good.<br>
><br>
><br>
> > If anyone is willing to take up this task, you can refer this paper:<br>
> > <a href="http://ols.108.redhat.com/2007/Reprints/briglia-Reprint.pdf" target="_blank">http://ols.108.redhat.com/2007/Reprints/briglia-Reprint.pdf</a><br>
> ><br>
> > This paper shows benchmarks of previous implementation of memory compression.<br>
> > It will give good idea on what to measure.<br>
><br>
> Are the scripts used to do that benchmark available? (I know<br>
> XAutomation is already available in Debian)<br>
><br>
<br>
<br>
</div>I am not much aware of testing tools they used. Maybe you can contact<br>
other authors of this paper for details.<br>
<font color="#888888"><br>
- Nitin<br>
</font><div><div></div><div class="Wj3C7c">_______________________________________________<br>
linux-mm-cc mailing list<br>
<a href="mailto:linux-mm-cc@lists.laptop.org">linux-mm-cc@lists.laptop.org</a><br>
<a href="http://lists.laptop.org/listinfo/linux-mm-cc" target="_blank">http://lists.laptop.org/listinfo/linux-mm-cc</a><br>
</div></div></blockquote></div><br>