<html><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">There are several MPI2007 SPEC submission using Itanium and x86-64. Here is a comparison between two submissions using 128 MPI processes, comparing an SGI Altix 4700 using Itanium 2 Processors 9040 1.6GHz and the fastest x86-64 submission which use Intel Xeon E5472, 3.00 GHz, IB 4x DDR, and Platform MPI 5.6.4. Both systems use 128 cores, whereas the Itanium system used 64 processors (chips) and the x86-64 system 32 processors (dual vs. quad-core).<div><br></div><div>Obviously, this is not apples-to-apples, since we here compare a large SMP (that is a single node) vs. a cluster of 16 dual quad-core (Harpertown) nodes connected by ConnectX. The x86-64 system is 15% faster than the Itanium system based on the SPECmpiM(TM)_base2007 metric (13.7 vs. 11.9). Here is a comparison on the individual benchmarks sorted on how many percent faster the x86-64 system is:<div><br class="webkit-block-placeholder"></div><div><br class="webkit-block-placeholder"></div><div apple-content-edited="true"> <span class="Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; orphans: 2; text-align: auto; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-border-horizontal-spacing: 0px; -webkit-border-vertical-spacing: 0px; -webkit-text-decorations-in-effect: none; -webkit-text-size-adjust: auto; -webkit-text-stroke-width: 0; "><div style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; "><div><table border="0" cellpadding="0" cellspacing="0" width="231" style="border-collapse:
collapse">
<!--StartFragment-->
<col width="89" style="mso-width-source:userset;mso-width-alt:3254">
<col width="142" style="mso-width-source:userset;mso-width-alt:5193">
<tbody><tr height="13">
<td height="13" width="89">Benchmark</td>
<td width="142">x86-64 % faster than ia64</td>
</tr>
<tr height="13">
<td height="13">107.leslie3d</td>
<td class="xl24" align="right" x:num="-0.552017937219731">-55.2%</td>
</tr>
<tr height="13">
<td height="13">137.lu</td>
<td class="xl24" align="right" x:num="-0.257575757575757">-25.8%</td>
</tr>
<tr height="13">
<td height="13">104.milc</td>
<td class="xl24" align="right" x:num="-0.236190476190476">-23.6%</td>
</tr>
<tr height="13">
<td height="13">130.socorro</td>
<td class="xl24" align="right" x:num="-0.103286384976526">-10.3%</td>
</tr>
<tr height="13">
<td height="13">127.wrf2</td>
<td class="xl24" align="right" x:num="0.0434782608695653">4.3%</td>
</tr>
<tr height="13">
<td height="13">128.GAPgeofem</td>
<td class="xl24" align="right" x:num="0.139369277721261">13.9%</td>
</tr>
<tr height="13">
<td height="13">132.zeusmp2</td>
<td class="xl24" align="right" x:num="0.154545454545454">15.5%</td>
</tr>
<tr height="13">
<td height="13">113.GemsFDTD</td>
<td class="xl24" align="right" x:num="0.167832167832168">16.8%</td>
</tr>
<tr height="13">
<td height="13">115.fds4</td>
<td class="xl24" align="right" x:num="0.23644251626898">23.6%</td>
</tr>
<tr height="13">
<td height="13">121.pop2</td>
<td class="xl24" align="right" x:num="0.623108665749656">62.3%</td>
</tr>
<tr height="13">
<td height="13">126.lammps</td>
<td class="xl24" align="right" x:num="0.736770691994573">73.7%</td>
</tr>
<tr height="13">
<td height="13">129.tera_tf</td>
<td class="xl24" align="right" x:num="0.753325272067715">75.3%</td>
</tr>
<tr height="13">
<td height="13">122.tachyon</td>
<td class="xl24" align="right" x:num="1.814685314685315">181.5%</td>
</tr>
<!--EndFragment-->
</tbody></table>
<br></div><div><br></div><div>The tendency here is clear, but not completely black-and-white. To me, this boils down to a YMMV discussion, essentially price performance for _my_ codes.</div><div><br></div><div>The complete data can be found here: <a href="http://www.spec.org/mpi2007/results/mpi2007.html">http://www.spec.org/mpi2007/results/mpi2007.html</a></div><div><br></div><div><br></div><div><br></div><div>Thanks, Håkon</div></div></span> </div><br><div><div><br></div></div></div></body></html>