tag:blogger.com,1999:blog-52625518346536757072011-04-22T04:40:57.083-07:00Derek Inglishttp://www.blogger.com/profile/12312901838031201613noreply@blogger.comBlogger41100tag:blogger.com,1999:blog-5262551834653675707.post-39855239272659802492009-05-08T09:46:00.000-07:002009-05-25T20:05:29.579-07:00Performance to Power Ratio score of <a href="ftp://ftp.software.ibm.com/eserver/benchmarks/news/newsblurb_dx360M2_specpower_052209.pdf">2066 overall ssj_ops/watt</a> * running IBM J9.<br /><br />* updated May 25, 2009 to point at updated IBM newblurb with new score<br /><br /><span style="font-size:78%;">SPEC is a registered trademark and SPECpower_ssj is a trademark of the Standard Performance Evaluation Corporation (see www.spec.org/spec/trademarks.html for all SPEC trademarks and service marks).</span><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/5262551834653675707-3985523927265980249?l=j9tr.blogspot.com' alt='' /></div>Derek Inglishttp://www.blogger.com/profile/12312901838031201613noreply@blogger.com1tag:blogger.com,1999:blog-5262551834653675707.post-43924741810318543812009-03-30T18:40:00.000-07:002009-03-30T21:31:03.153-07:00It's been fun following <a href="http://www.intc.com/eventdetail.cfm?EventID=67088">Intel's launch of the 2-socket Nehalem chip.</a> This is the first time that I can remember having SPECjbb2005 results available on launch day using each of the 'big 3' JVMs.<br /><br />Kudos to Oracle JRockit for having the <a href="http://spec.org/jbb2005/results/res2009q1/jbb2005-20090310-00671.html">first Nehalem based SPECjbb2005</a> results available on www.spec.org using a new P28.0.0. JVM.<br /><br />The Sun HotSpot results using a new Java 6 Update 14 Performance Release are referenced <a href="http://blogs.sun.com/dagastine/entry/java_rocks_intel_nehalem1">here.</a><br /><br /><span style="font-weight: bold;">The <a href="ftp://ftp.software.ibm.com/eserver/benchmarks/news/newsblurb_HS22_specjbb_033009.pdf">IBM J9 result</a> is 8-10% faster than the JRockit and HotSpot results</span> .... not surprisingly Intel went with the J9 result on their <a href="http://www.intel.com/performance/server/xeon/server.htm?iid=perf_server_lhn+dp_server">Nehalem benchmark page</a>.<br /><a onblur="try {parent.deselectBloggerImageGracefully();} catch(e) {}" href="http://2.bp.blogspot.com/_tpkqcOJGWDg/SdGaluWU0RI/AAAAAAAAADA/2gsJYxSbvhU/s1600-h/Nehalem.png"><img style="margin: 0px auto 10px; display: block; text-align: center; cursor: pointer; width: 400px; height: 254px;" src="http://2.bp.blogspot.com/_tpkqcOJGWDg/SdGaluWU0RI/AAAAAAAAADA/2gsJYxSbvhU/s400/Nehalem.png" alt="" id="BLOGGER_PHOTO_ID_5319202607788773650" border="0" /></a>Derek<br /><br />The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.<br /><span style="font-size:85%;"><br /><span class="textsmall">SPEC and SPECjbb are registered trademarks of the Standard Performance Evaluation Corporation. </span></span><br /><span style="font-size:85%;">Results referenced are current as of March 30, 2009. The SPECjbb2005 results have been submitted to SPEC for review. Upon successful review, the result will be posted at <a href="http://www.spec.org/">www.spec.org</a>, which contains a complete list of published SPECjbb2005 results. SPEC, SPECjbb reg tm of Standard Performance Evaluation Corporation.<br /><br /><span style="font-size:78%;"><a href="ftp://ftp.software.ibm.com/eserver/benchmarks/news/newsblurb_HS22_specjbb_033009.pdf">IBM BladeCenter HS22 (2 chips, 8 cores) SPECjbb2005 bops = 604,417, SPECjbb2005 bops/JVM = 151,104.</a><br /></span></span><span style="font-size:78%;"><a href="http://blogs.sun.com/dagastine/entry/java_rocks_intel_nehalem1">Sun Branded Intel Xeon X5570 System (2 chips, 8 cores) SPECjbb2005 bops = 556,882, SPECjbb2005 bops/JVM = 278,441.</a><br /></span><span style="font-size:85%;"><a href="http://spec.org/jbb2005/results/res2009q1/jbb2005-20090310-00671.html"><span style="font-size:78%;">PowerEdge M710 (2 chips, 8 cores) SPECjbb2005 bops = 549,197, SPECjbb2005 bops/JVM = 274,599.</span></a><br /></span><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/5262551834653675707-4392474181031854381?l=j9tr.blogspot.com' alt='' /></div>Derek Inglishttp://www.blogger.com/profile/12312901838031201613noreply@blogger.com3tag:blogger.com,1999:blog-5262551834653675707.post-87775142872604833242008-12-11T19:24:00.000-08:002008-12-11T22:09:53.318-08:00An homage to John "cool man" <span class="blsp-spelling-error" id="SPELLING_ERROR_0">Stecher</span>.<br /><br />When everyone knows you're the fastest, you can play it cool.<br /><br /><a href="http://www-01.ibm.com/software/webservers/appserv/was/"><span class="blsp-spelling-error" id="SPELLING_ERROR_1">WebSphere</span> Application Server V7.0</a> <span class="blsp-spelling-error" id="SPELLING_ERROR_2">GA'd</span> in October. A couple weeks later IBM publishes an overall leadership <a href="http://www.spec.org/osg/jAppServer2004/results/res2008q4/jAppServer2004-20081022-00121.html"><span class="blsp-spelling-error" id="SPELLING_ERROR_3">SPECjAppServer</span>2004 score of 20179 JOPS@Standard.</a> Now, a month later, if you google "<span class="blsp-spelling-error" id="SPELLING_ERROR_4">SPECjAppServer</span>2004 20179" you won't find a single hit ... well, maybe now you will. <span class="blsp-spelling-error" id="SPELLING_ERROR_5">Hm, a little more digging and I see that </span>SPEC uses different score formats depending on how the data is queried. The score is also reported as 20178.61 JOPS@Standard which does give me a couple of hits.<br /><br />Damn, I thought WAS was going for the ultimate cool, man.<br /><br />Derek "never going to be a cool man" Inglis<br /><br /><span style="font-size:85%;">SPEC and SPECjAppServer are trademarks or registered trademarks of the Standard Performance Evaluation Corporation. </span><span style="font-size:85%;">Results from <a href="http://www.spec.org/">www.spec.org</a> as of December 11th 2008.</span><br /><span style="font-size:85%;"><br />The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.</span><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/5262551834653675707-8777514287260483324?l=j9tr.blogspot.com' alt='' /></div>Derek Inglishttp://www.blogger.com/profile/12312901838031201613noreply@blogger.com0tag:blogger.com,1999:blog-5262551834653675707.post-43957538200770128542008-12-10T11:04:00.000-08:002008-12-11T05:24:43.706-08:00With the release of <a href="http://www.ibm.com/developerworks/java/jdk/linux/download.html">IBM Java 6 SR3</a>, and its availablility on Windows via <a href="http://www-01.ibm.com/support/docview.wss?rs=180&amp;uid=swg24021068">Websphere Application Server 7.0.0.1</a>, J9 has extended its SPECjbb2005 lead over JRockit to 11% on Intel and has trumped its own leading result on AMD by 10%.<br /><br />IBM J9 has<br /><ul><li>leadership <a href="http://www.spec.org/osg/jbb2005/results/res2007q2/jbb2005-20070405-00284.html">1</a>,<a href="http://www.spec.org/osg/jbb2005/results/res2007q2/jbb2005-20070519-00309.html">2</a>,<a href="http://www.spec.org/osg/jbb2005/results/res2008q2/jbb2005-20080604-00502.html">4</a>,<a href="http://www.spec.org/osg/jbb2005/results/res2008q2/jbb2005-20080604-00503.html">8</a>,<a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081006-00534.html">16</a>,<a href="http://www.spec.org/osg/jbb2005/results/res2008q3/jbb2005-20080826-00525.html">32</a> &amp; <a href="http://www.spec.org/osg/jbb2005/results/res2008q2/jbb2005-20080407-00477.html">64</a>-core results (<a href="http://www-03.ibm.com/systems/p/">IBM System p</a> and <a href="http://www-03.ibm.com/systems/i/">System i</a>)</li><li>leadership <a href="http://www.spec.org/osg/jbb2005/results/res2008q3/jbb2005-20080815-00519.html">4</a> and <a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081031-00555.html">8</a>-core Intel results</li><li>leadership <a href="ftp://ftp.software.ibm.com/eserver/benchmarks/news/newsblurb_LS42-2P_specjbb_120208.pdf">8</a> and <a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081112-00559.html">16</a>-core AMD results</li><li><a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081112-00559.html">leadership 4-socket result</a></li><li><a href="ftp://ftp.software.ibm.com/eserver/benchmarks/news/newsblurb_LS42-2P_specjbb_120208.pdf">leadership x86 2-socket result</a><br /></li></ul><br />Published results on Intel's Harpertown chip provide an interesting comparison of J9, JRockit and HotSpot. (Note the HotSpot number is using Sun's Java 6 _05 update. There hasn't been a publish with their faster _06 Performance update.)<br /><a onblur="try {parent.deselectBloggerImageGracefully();} catch(e) {}" href="http://4.bp.blogspot.com/_tpkqcOJGWDg/SUAaIKpdLNI/AAAAAAAAABQ/QtEV4W9-B5A/s1600-h/jbb1.png"><img style="margin: 0px auto 10px; display: block; text-align: center; cursor: pointer; width: 401px; height: 252px;" src="http://4.bp.blogspot.com/_tpkqcOJGWDg/SUAaIKpdLNI/AAAAAAAAABQ/QtEV4W9-B5A/s400/jbb1.png" alt="" id="BLOGGER_PHOTO_ID_5278247490877730002" border="0" /></a><br />J9's SR1 publish on the 3.16GHz Harpertown chip beats the more recent JRockit publish on the 3.33GHz chip. Also of interest are the uses of 32 and 64-bit JVMs. J9's 64-bit SR3 number beats JRockit by 11% and J9's 32-bit number beats JRockit by 6%.<br /><br />J9 now has the leadership 8-core Intel number, the leadership 8-core AMD number and the overall leadership 8-core number using an <a href="http://www-03.ibm.com/systems/power/advantages/power6/index.html">IBM POWER6</a> chip.<br /><a onblur="try {parent.deselectBloggerImageGracefully();} catch(e) {}" href="http://2.bp.blogspot.com/_tpkqcOJGWDg/SUAcRGDXHHI/AAAAAAAAABo/gmzIkqzWi4k/s1600-h/jbb2.png"><img style="margin: 0px auto 10px; display: block; text-align: center; cursor: pointer; width: 400px; height: 296px;" src="http://2.bp.blogspot.com/_tpkqcOJGWDg/SUAcRGDXHHI/AAAAAAAAABo/gmzIkqzWi4k/s400/jbb2.png" alt="" id="BLOGGER_PHOTO_ID_5278249843286285426" border="0" /></a>With Sun's Niagara-2 processors having 8 cores per socket, Sun has published a number of leading per-socket scores. With the recent publish on <a href="ftp://ftp.software.ibm.com/eserver/benchmarks/news/newsblurb_LS42_specjbb_111308.pdf">IBM's Bladecenter LS42</a>, IBM and J9 now claim the leadership 4-socket result:<a onblur="try {parent.deselectBloggerImageGracefully();} catch(e) {}" href="http://4.bp.blogspot.com/_tpkqcOJGWDg/SUAf15p6tzI/AAAAAAAAACA/OnjBOCf324E/s1600-h/jbb5.png"><img style="margin: 0px auto 10px; display: block; text-align: center; cursor: pointer; width: 400px; height: 279px;" src="http://4.bp.blogspot.com/_tpkqcOJGWDg/SUAf15p6tzI/AAAAAAAAACA/OnjBOCf324E/s400/jbb5.png" alt="" id="BLOGGER_PHOTO_ID_5278253774148384562" border="0" /></a>Sun is still leading the 2-socket category ... by 2% ... using twice the number of cores and 16 times the number of threads.<br /><a onblur="try {parent.deselectBloggerImageGracefully();} catch(e) {}" href="http://4.bp.blogspot.com/_tpkqcOJGWDg/SUAcaAQRBII/AAAAAAAAABw/ExZyRwrHiLg/s1600-h/jbb3.png"><img style="margin: 0px auto 10px; display: block; text-align: center; cursor: pointer; width: 400px; height: 298px;" src="http://4.bp.blogspot.com/_tpkqcOJGWDg/SUAcaAQRBII/AAAAAAAAABw/ExZyRwrHiLg/s400/jbb3.png" alt="" id="BLOGGER_PHOTO_ID_5278249996348621954" border="0" /></a>J9 has the overal leadership 16-core result as well as the leading x86 16-core result.<br /><a onblur="try {parent.deselectBloggerImageGracefully();} catch(e) {}" href="http://3.bp.blogspot.com/_tpkqcOJGWDg/SUAcinJSPeI/AAAAAAAAAB4/LnmGT8_kppE/s1600-h/jbb4.png"><img style="margin: 0px auto 10px; display: block; text-align: center; cursor: pointer; width: 400px; height: 287px;" src="http://3.bp.blogspot.com/_tpkqcOJGWDg/SUAcinJSPeI/AAAAAAAAAB4/LnmGT8_kppE/s400/jbb4.png" alt="" id="BLOGGER_PHOTO_ID_5278250144227278306" border="0" /></a><br />oh .. if you actually read this far and are wondering what we did in SR3. The 10% boost is from the <a href="http://harmony.apache.org/">Apache Harmony</a> improvement to TreeMap discussed in <a href="http://blogs.sun.com/dagastine/entry/apache_harmony_thanks_for_the">David Dagastine's blog</a>.<br /><br />Derek<br /><br />The postings on this site are my own and don’t necessarily represent IBM’s positions, strategies or opinions.<br /><span style="font-size:85%;"><br /><span class="textsmall">SPEC and SPECjbb are registered trademarks of the Standard Performance Evaluation Corporation. </span></span><br /><span style="font-size:85%;">Results referenced are current as of December 10, 2008. The SPECjbb2005 results have been submitted to SPEC for review. Upon successful review, the result will be posted at <a href="http://www.spec.org/">www.spec.org</a>, which contains a complete list of published SPECjbb2005 results. SPEC, SPECjbb reg tm of Standard Performance Evaluation Corporation.<br /></span><span style="font-size:85%;"><span style="font-size:78%;"><br />1-core<br /><a href="http://www.spec.org/osg/jbb2005/results/res2007q2/jbb2005-20070405-00284.html">IBM System i5 515 (1 chip, 1 core, 2 threads) SPECjbb2005 bops = 20615, SPECjbb2005 bops/JVM=20615</a><br /><br />2-core<br /><a href="http://www.spec.org/osg/jbb2005/results/res2007q2/jbb2005-20070519-00309.html">IBM System p 570 (1 chips, 2 cores, 4 thredas) SPECjbb2005 bops=88089, SPECjbb2005 bops/JVM=88089</a><br /><br />4-core<br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q2/jbb2005-20080604-00502.html">IBM Power 570 (2 chips, 4 cores, 8 threads) SPECjbb2005 bops=205917, SPECjbb2005 bops/JVM=102959</a><br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q3/jbb2005-20080815-00519.html">IBM System X3350 (1 chip, 4 cores, 4 threads) SPECjbb2005 bops = 194256, SPECjbb2005 bops/JVM = 97128</a><br /><br />8-core<br /></span></span><span style="font-size:85%;"><span style="font-size:78%;"><a href="http://www.spec.org/osg/jbb2005/results/res2008q2/jbb2005-20080604-00503.html">IBM Power 570 (4 chips, 8 cores, 16 threads) SPECjbb2005 bops = 402.923, SPECjbb2005 bops/JVM = 100731. </a><br /><a href="ftp://ftp.software.ibm.com/eserver/benchmarks/news/newsblurb_LS42-2P_specjbb_120208.pdf">IBM Bladecenter LS42 (2 chips, 8 cores, 8 threads) SPECjbb2005 bops=382,339, SPECjbb2005 bops/JVM=191,170</a><br /></span></span><span style="font-size:85%;"><span style="font-size:78%;"><a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081031-00555.html">IBM System x3650 (2 chips, 8 cores, 8 threads) SPECjbb2005 bops = 344,436, SPECjbb2005 bops/JVM = 86,109.</a><br /></span></span><a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081031-00556.html"><span style="font-size:85%;"><span style="font-size:78%;">IBM System x3650 (2 chips, 8 cores, 8 threads) SPECjbb2005 bops = 334,802, SPECjbb2005 bops/JVM = 83,701.</span></span></a><br /><span style="font-size:85%;"><span style="font-size:78%;"><a href="http://www.spec.org/osg/jbb2005/results/res2008q1/jbb2005-20080311-00460.html">IBM System x3650 (2 chips, 8 cores, 8 thredas) SPECjbb2005 bops = 323172, SPECjbb2005 bops/JVM = 80793.</a><br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q3/jbb2005-20080826-00522.html">PRIMERGY RX200 S4 (2 chips, 8 cores, 8 threads,) SPECjbb2005 bops = 316,728, SPECjbb2005 bops/JVM = 79,182. </a><br /></span></span><a href="http://www.spec.org/osg/jbb2005/results/res2008q1/jbb2005-20080208-00452.html"><span style="font-size:85%;"><span style="font-size:78%;">Sun Fire X4150 (2 chips, 8 cores, 8 threads) SPECjbb2005 bops = 303297, SPECjbb2005 bops/JVM = 75,824. </span></span><br /></a><span style="font-size:85%;"><span style="font-size:78%;"><br />16-core<br /></span></span><a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081006-00534.html"><span style="font-size:85%;"><span style="font-size:78%;">IBM Power 570 (8 chips, 16 cores, 32 threads) SPECjbb2005 bops = 867989, SPECjbb2005 bops/JVM = 108499.<br /></span></span></a><span style="font-size:85%;"><span style="font-size:78%;"><a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081112-00559.html">IBM Bladecenter LS42 (4 chips, 16 cores, 16 threads) SPECjbb2005 bops = 721843, SPECjbb2005 bops/JVM = 180461.</a><br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081024-00550.html">Transport TX46 (4 chips, 16 cores, 16 threads) SPECjbb2005 bops = 659034, SPECjbb2005 bops/JVM = 164759.</a><br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q2/jbb2005-20080506-00485.html">Sun Fire X4450 (4 chips, 16 cores, 16 threads) SPECjbb2005 bops = 464355, SPECjbb2005 bops/JVM = 58044.</a></span></span><br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081021-00539.html"><span style="font-size:78%;">Sun Blade T6340 (2 chips, 16 cores, 128 threads) SPECjbb2005 bops = 388456, SPECjbb2005 bops/JVM = 24279.</span></a><br /><span style="font-size:85%;"><span style="font-size:78%;"></span></span><span style="font-size:78%;"><br />24-core<br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q3/jbb2005-20080819-00520.html">Sun Fire X4450 (4 chips, 24 cores, 24 threads) SPECjbb2005 bops = 531669, SPECjbb2005 bops/JVM = 132917.</a><br /><br />32-core<a href="http://www.spec.org/osg/jbb2005/results/res2008q3/jbb2005-20080826-00525.html"><br />IBM Power 595 (16 chips, 32 cores, 64 threads) SPECjbb2005 bops = 1530343, SPECjbb2005 bops/JVM = 95646.</a><br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q4/jbb2005-20081007-00536.html">Sun SPARC Enterprise T5440 (4 chips, 32 cores, 256 threads) SPECjbb2005 bops = 692736, SPECjbb2005 bops/JVM = 21648.</a><br /><br />64-core<br /><a href="http://www.spec.org/osg/jbb2005/results/res2008q2/jbb2005-20080407-00477.html">IBM Power 585 (32 chips, 64 cores, 128 threads) SPECjbb2005 bops = 3435485, SPECjbb2005 bops/JVM = 107359.</a><br /></span><div class="blogger-post-footer"><img width='1' height='1' src='https://blogger.googleusercontent.com/tracker/5262551834653675707-4395753820077012854?l=j9tr.blogspot.com' alt='' /></div>Derek Inglishttp://www.blogger.com/profile/12312901838031201613noreply@blogger.com2