{"id":866,"date":"2017-04-05T14:33:01","date_gmt":"2017-04-05T12:33:01","guid":{"rendered":"https:\/\/webs.uab.cat\/hpc4eas\/news\/conference-at-ibm-new-york\/"},"modified":"2017-04-05T14:33:01","modified_gmt":"2017-04-05T12:33:01","slug":"conference-at-ibm-new-york","status":"publish","type":"news","link":"https:\/\/webs.uab.cat\/hpc4eas\/news\/conference-at-ibm-new-york\/","title":{"rendered":"Conference at IBM (New York)."},"content":{"rendered":"<p><strong>Abstract<\/strong><\/p>\n<div>In parallel applications executed in HPC environments there are idle times during the execution that can be caused by communication collisions or cache conflits generating cache misses.&nbsp;<\/div>\n<div>We are developing a methodology, for message passing parallel scientific applications, which attempts to reduce the applications execution time by reducing these idle times.<br \/>\nIn our proposal, to allocate processes to processors, we define two kind of &#8220;forces&#8221; acting among the processes: attraction and repulsion forces.<\/div>\n<div>\n<ol>\n<li><strong>Attraction forces<\/strong> are used to approximate those processes interchaning high volumes of communications, in order to speed up the communication.<\/li>\n<li><strong>Repulsion forces<\/strong> are used to &#8220;separate&#8221; those processes with high volumes of data, in order to avoid they can share common caches, generating high cache misses ratios.<\/li>\n<\/ol>\n<p><img loading=\"lazy\" decoding=\"async\" class=\" size-full wp-image-864\" alt=\"\" height=\"159\" src=\"https:\/\/webs.uab.cat\/hpc4eas\/wp-content\/uploads\/sites\/289\/2017\/04\/ibm2016.jpg\" width=\"320\" srcset=\"https:\/\/webs.uab.cat\/hpc4eas\/wp-content\/uploads\/sites\/289\/2017\/04\/ibm2016.jpg 320w, https:\/\/webs.uab.cat\/hpc4eas\/wp-content\/uploads\/sites\/289\/2017\/04\/ibm2016-300x149.jpg 300w\" sizes=\"auto, (max-width: 320px) 100vw, 320px\" \/><\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\" size-full wp-image-865\" alt=\"\" height=\"178\" src=\"https:\/\/webs.uab.cat\/hpc4eas\/wp-content\/uploads\/sites\/289\/2017\/04\/ibm2016-2.jpg\" width=\"190\" \/><\/p>\n<\/div>\n","protected":false},"excerpt":{"rendered":"<p>Abstract In parallel applications executed in HPC environments there are idle times during the execution that can be caused by communication collisions or cache conflits generating cache misses.&nbsp; We are developing a methodology, for message passing parallel scientific applications, which attempts to reduce the applications execution time by reducing these idle times. In our proposal, [&hellip;]<\/p>\n","protected":false},"author":20,"featured_media":864,"template":"","class_list":["post-866","news","type-news","status-publish","has-post-thumbnail","hentry"],"_links":{"self":[{"href":"https:\/\/webs.uab.cat\/hpc4eas\/wp-json\/wp\/v2\/news\/866","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/webs.uab.cat\/hpc4eas\/wp-json\/wp\/v2\/news"}],"about":[{"href":"https:\/\/webs.uab.cat\/hpc4eas\/wp-json\/wp\/v2\/types\/news"}],"author":[{"embeddable":true,"href":"https:\/\/webs.uab.cat\/hpc4eas\/wp-json\/wp\/v2\/users\/20"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/webs.uab.cat\/hpc4eas\/wp-json\/wp\/v2\/media\/864"}],"wp:attachment":[{"href":"https:\/\/webs.uab.cat\/hpc4eas\/wp-json\/wp\/v2\/media?parent=866"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}