{"id":50732,"date":"2019-09-10T21:23:23","date_gmt":"2019-09-10T19:23:23","guid":{"rendered":"https:\/\/www.planet3dnow.de\/cms\/?p=50732"},"modified":"2019-09-10T21:23:23","modified_gmt":"2019-09-10T19:23:23","slug":"exasol-and-amd-achieve-world-record-results-on-tpc-h-benchmark","status":"publish","type":"post","link":"https:\/\/www.planet3dnow.de\/cms\/50732-exasol-and-amd-achieve-world-record-results-on-tpc-h-benchmark\/","title":{"rendered":"Exasol and AMD<\/span> Achieve World Record Results on TPC<\/span>\u2011H Benchmark"},"content":{"rendered":"

\u201c<\/span>Best ever\u201d results for a 1 TB<\/span> scale factor on industry standard decision support benchmark were achieved with AMD<\/span> EPYC<\/span>\u2122 7002 series processors<\/em><\/h3>\n

 <\/p>\n

Lon\u00addon, 10 Sep\u00adtem\u00adber 2019<\/strong> \u2013 Exa\u00adsol, the ana\u00adly\u00adtics data\u00adba\u00adse, is proud to announ\u00adce world record \u201cbest ever\u201d results on indus\u00adtry stan\u00addard decis\u00adi\u00adon sup\u00adport bench\u00admark TPC<\/span>\u2011H. Exa\u00adsol has main\u00adtai\u00adned its posi\u00adti\u00adon as the undis\u00adpu\u00adted lea\u00adder \u2013 by a signi\u00adfi\u00adcant mar\u00adgin \u2013 for both raw per\u00adfor\u00admance and pri\u00adce-per\u00adfor\u00admance for over a deca\u00adde across seve\u00adral cate\u00adgo\u00adries. In July, the com\u00adpa\u00adny announ\u00adced out\u00adstan\u00adding results for 3 TB<\/span> and 10 TB<\/span> sca\u00adle fac\u00adtors. Now, run\u00adning the bench\u00admark on AMD<\/span> EPYC<\/span>\u2122 7002 series pro\u00adces\u00adsor-based ser\u00advers, records have been bro\u00adken for the 1 TB<\/span> sca\u00adle fac\u00adtor category.<\/p>\n

\u201c<\/span>We are very exci\u00adted to be working with AMD<\/span> on new and more ambi\u00adtious bench\u00admarks to put both Exa\u00adsol and AMD<\/span> EPYC<\/span> pro\u00adces\u00adsor-based ser\u00advers to the test. The results we have achie\u00adved with the new gene\u00adra\u00adti\u00adon pro\u00adces\u00adsors are the best ever for this cate\u00adgo\u00adry in par\u00adti\u00adcu\u00adlar,\u201d said Mathi\u00adas Golom\u00adbek, CTO<\/span> at Exasol.<\/p>\n

The bench\u00admark was com\u00adple\u00adted on a sca\u00adle fac\u00adtor of 1 Tera\u00adbyte (TB<\/span>) on 4 nodes and 8 nodes, demons\u00adt\u00adra\u00adting out\u00adstan\u00adding sca\u00adla\u00adbi\u00adli\u00adty and brea\u00adking records for per\u00adfor\u00admance and pri\u00adce-per\u00adfor\u00admance. At over 6.1 mil\u00adli\u00adon QphH@1TB<\/span>, this is the best per\u00adfor\u00admance ever recor\u00added at the 1 TB<\/span> sca\u00adle fac\u00adtor. The pri\u00adce per per\u00adfor\u00admance result, also the best ever recor\u00added for this cate\u00adgo\u00adry, is $0.06.<\/p>\n

The\u00adse results, available<\/a> as of 31 July 2019, demons\u00adtra\u00adte Exasol\u2019s speed, sca\u00adla\u00adbi\u00adli\u00adty and cost\/performance which have con\u00adsis\u00adt\u00adent\u00adly domi\u00adna\u00adted the TPC<\/span>\u2011H bench\u00admarks<\/a> sin\u00adce 2008. The latest results were achie\u00adved with Exa\u00adsol 6.2 run\u00adning on a clus\u00adter of eight HPE<\/span> Pro\u00adLi\u00adant DL325<\/span> Gen10 ser\u00advers, each with one AMD<\/span> EPYC<\/span>TM<\/span><\/sup> 7702 64 core processor.<\/p>\n

\u201c<\/span>We are plea\u00adsed that Exa\u00adsol has once more achie\u00adved remar\u00adkab\u00adle bench\u00admark results, this time using 2nd<\/sup> Gene\u00adra\u00adti\u00adon AMD<\/span> EPYC<\/span> pro\u00adces\u00adsor-based ser\u00advers,\u201d said Rag\u00adhu Nam\u00adbi\u00adar, cor\u00adpo\u00adra\u00adte vice pre\u00adsi\u00addent and CTO<\/span>, Dat\u00ada\u00adcen\u00adter Eco\u00adsys\u00adtems &<\/span> Appli\u00adca\u00adti\u00adon Engi\u00adnee\u00adring, AMD<\/span>. \u201cPer\u00adfor\u00admance demands are incre\u00adasing but IT<\/span> bud\u00adgets and data cen\u00adter space are not. Built on 7nm tech\u00adno\u00adlo\u00adgy, AMD<\/span> EPYC<\/span>\u2122 7002 Series pro\u00adces\u00adsors pack more cores and more capa\u00adbi\u00adli\u00adty into the same space as the pre\u00advious gene\u00adra\u00adti\u00adon, deli\u00adve\u00adring excep\u00adtio\u00adnal per\u00adfor\u00admance and scalability\u201d<\/p>\n

Over the years, the\u00adre have been over 220 recor\u00added TPC<\/span>\u2011H bench\u00admarks. Many data\u00adba\u00adse ven\u00addors have pos\u00adted results on a vast ran\u00adge of hard\u00adware and at various sca\u00adle fac\u00adtors up to 100TB<\/span>. By run\u00adning the same unbi\u00ada\u00adsed scripts, you can compa\u00adre data\u00adba\u00adse ven\u00addor with data\u00adba\u00adse ven\u00addor \u2013 and you can often see how well the same data\u00adba\u00adse runs on dif\u00adfe\u00adrent hardware.<\/p>\n

 <\/p>\n

Notes:<\/strong><\/p>\n

The full TPC<\/span>\u2011H bench\u00admark results are as fol\u00adlows and are available as of 07\/31\/19. Full results can be found in the link pro\u00advi\u00added below.<\/p>\n

SF<\/span> 1000
\nAvai\u00adla\u00adbi\u00adli\u00adty Date: 07\/31\/19
\nPer\u00adfor\u00admance: 6,145,628 QphH@1000GB<\/span>
\nPrice\/Performance: 0.06 USD<\/span> per QphH@1000GB<\/span>
\nSys\u00adtem: 8x HPE<\/span> Pro\u00adLi\u00adant DL325<\/span> Gen10 ser\u00advers, each with:<\/p>\n

256 GB<\/span> RAM<\/span>, 1x AMD<\/span> EPYCTM<\/span> 7702, 1x 1.6TB<\/span> NVMe<\/p>\n

Refe\u00adrence URL<\/span>: http:\/\/www.tpc.org\/3341<\/a>
\nClaims Cur\u00adrent of: 07\/31\/19<\/p>\n

 <\/p>\n

About the TPC<\/span>\u2011H benchmarks<\/strong><\/h4>\n

The TPC<\/span>\u2011H bench\u00admark is a per\u00adfor\u00admance test of the TPC<\/span>, a non-pro\u00adfit cor\u00adpo\u00adra\u00adti\u00adon foun\u00added to defi\u00adne data\u00adba\u00adse bench\u00admarks and to dis\u00adse\u00admi\u00adna\u00adte objec\u00adti\u00adve, veri\u00adfia\u00adble per\u00adfor\u00admance data to the indus\u00adtry. The TPC<\/span>\u2011H bench\u00admark is a decis\u00adi\u00adon sup\u00adport bench\u00admark; it con\u00adsists of a suite of busi\u00adness-ori\u00aden\u00adted ad-hoc queries and con\u00adcur\u00adrent data modi\u00adfi\u00adca\u00adti\u00adons. The queries and the data popu\u00adla\u00adting the data\u00adba\u00adse have been cho\u00adsen to have broad indus\u00adtry-wide rele\u00advan\u00adce. This bench\u00admark illus\u00adtra\u00adtes decis\u00adi\u00adon sup\u00adport sys\u00adtems that exami\u00adne lar\u00adge volu\u00admes of data, exe\u00adcu\u00adte queries with a high degree of com\u00adple\u00adxi\u00adty, and give ans\u00adwers to cri\u00adti\u00adcal busi\u00adness questions.<\/p>\n

 <\/p>\n

About Exasol<\/strong><\/h4>\n

Exa\u00adsol is the ana\u00adly\u00adtics data\u00adba\u00adse. Its high-per\u00adfor\u00admance in-memo\u00adry ana\u00adly\u00adtics data\u00adba\u00adse gives orga\u00adniza\u00adti\u00adons the power to trans\u00adform how they work with data, on-pre\u00admi\u00adses, in the cloud or both \u2013 and turn it into value fas\u00adter, easier and more cost effec\u00adtively than ever before.<\/p>\n

To learn more about Exa\u00adsol plea\u00adse visit www.exasol.com<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

\u201c<\/span>Best ever\u201d results for a 1 TB<\/span> sca\u00adle fac\u00adtor on indus\u00adtry stan\u00addard decis\u00adi\u00adon sup\u00adport bench\u00admark were achie\u00adved with AMD<\/span> EPYC<\/span>\u2122 7002 series pro\u00adces\u00adsors   Lon\u00addon, 10 Sep\u00adtem\u00adber 2019 \u2013 Exa\u00adsol, the ana\u00adly\u00adtics data\u00adba\u00adse, is proud to announ\u00adce world record \u201cbest ever\u201d results on indus\u00adtry stan\u00addard decis\u00adi\u00adon sup\u00adport bench\u00admark TPC<\/span>\u2011H. Exa\u00adsol has main\u00adtai\u00adned its posi\u00adti\u00adon as the (\u2026) Wei\u00adter\u00adle\u00adsen \u00bb<\/a><\/p>\n","protected":false},"author":593,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"wp_typography_post_enhancements_disabled":false,"ngg_post_thumbnail":0,"footnotes":""},"categories":[22],"tags":[966,532,1983,2052],"class_list":["post-50732","post","type-post","status-publish","format-standard","hentry","category-pressemitteilungen","tag-amd","tag-benchmark","tag-exasol","tag-world-record","entry"],"share_on_mastodon":{"url":"","error":""},"_links":{"self":[{"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/posts\/50732","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/users\/593"}],"replies":[{"embeddable":true,"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/comments?post=50732"}],"version-history":[{"count":1,"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/posts\/50732\/revisions"}],"predecessor-version":[{"id":50733,"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/posts\/50732\/revisions\/50733"}],"wp:attachment":[{"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/media?parent=50732"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/categories?post=50732"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.planet3dnow.de\/cms\/wp-json\/wp\/v2\/tags?post=50732"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}