<br><font size=2 face="sans-serif">Alex,</font>
<br>
<br><font size=2 face="sans-serif">There is no formal test suite. However
any application/workload can used to generate an overcommit situation.
The three scenarios of interest are:</font>
<br>
<ol>
<li value=1><font size=2 face="sans-serif">Two or more applications/workloads
are run simultaneously, each of which does not overcommit the SPEs. </font>
<li value=2><font size=2 face="sans-serif">One application/workloads that
overcommits the SPEs. Many of our code samples allows one to specify the
number of SPEs.</font>
<li value=3><font size=2 face="sans-serif">Combination of 1 and 2.</font></ol>
<br><font size=2 face="sans-serif">#1 is probably the most important scenario.
</font>
<br><font size=2 face="sans-serif"><br>
Sincerely,<br>
<br>
Dan B.<br>
-------------------------------------------------------------------------------<br>
Daniel A. Brokenshire<br>
STSM / Emerging Products Development<br>
brokensh@us.ibm.com 512-838-3373</font>
<br>
<br>
<br>
<table width=100%>
<tr valign=top>
<td width=40%><font size=1 face="sans-serif"><b>Alex Chow/Austin/IBM@IBMUS</b>
</font>
<br><font size=1 face="sans-serif">Sent by: cbe-oss-dev-bounces+brokensh=us.ibm.com@ozlabs.org</font>
<p><font size=1 face="sans-serif">10/24/2007 08:33 AM</font>
<td width=59%>
<table width=100%>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">To</font></div>
<td><font size=1 face="sans-serif">Christoph Hellwig <hch@lst.de></font>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">cc</font></div>
<td><font size=1 face="sans-serif">Kanna Shimizu/Austin/IBM@IBMUS, Arnd
Bergmann <arnd.bergmann@de.ibm.com>, cbe-oss-dev@ozlabs.org, Christoph
Hellwig <hch@lst.de>, cbe-oss-dev-bounces+achow=us.ibm.com@ozlabs.org</font>
<tr valign=top>
<td>
<div align=right><font size=1 face="sans-serif">Subject</font></div>
<td><font size=1 face="sans-serif">Re: [Cbe-oss-dev] PATCH [0/3] spusched
enhancements</font></table>
<br>
<table>
<tr valign=top>
<td>
<td></table>
<br></table>
<br>
<br>
<br><font size=3>Hi, is there a performance test suite to measure the overhead
impact of "over committed" case?<br>
<br>
Thanks..<br>
=======<br>
Regards,<br>
<br>
Alex Chunghen Chow<br>
<br>
</font><tt><font size=3><br>
cbe-oss-dev-bounces+achow=us.ibm.com@ozlabs.org wrote on 10/24/2007 03:23:32
AM:<br>
<br>
> Hi Luke,<br>
> <br>
> I like this a lot. I remember we had some problems with pagefault<br>
> handling with this idea, so we might have to look into that again.<br>
> <br>
> More detailed comments later.<br>
> <br>
> On Tue, Oct 23, 2007 at 11:09:02PM -0300, Luke Browning wrote:<br>
> > Hi Arnd and Christoph,<br>
> > <br>
> > This patchset should significantly improve the performance of
the system<br>
> > when the SPUs are over committed. I have tested these changes
on the<br>
> > simulator with good results. I can now run several instances
of the<br>
> > matrix multiplication testcase at the same time in a few minutes.
<br>
> > <br>
> > I haven't tested 'isolated' spu contexts yet. <br>
> > <br>
> > Do you have a testcase for 'isolated' contexts?<br>
> <br>
> I think Jeremy's testsuite now has a testcase for it.<br>
> <br>
> _______________________________________________<br>
> cbe-oss-dev mailing list<br>
> cbe-oss-dev@ozlabs.org<br>
> </font></tt><a href="https://ozlabs.org/mailman/listinfo/cbe-oss-dev"><tt><font size=3 color=blue><u>https://ozlabs.org/mailman/listinfo/cbe-oss-dev</u></font></tt></a><tt><font size=2>_______________________________________________<br>
cbe-oss-dev mailing list<br>
cbe-oss-dev@ozlabs.org<br>
https://ozlabs.org/mailman/listinfo/cbe-oss-dev<br>
</font></tt>
<br>