{"id":50,"date":"2005-09-13T21:12:33","date_gmt":"2005-09-14T02:12:33","guid":{"rendered":"http:\/\/blogs.law.harvard.edu\/hoanga\/2005\/09\/13\/gpgpu-and-oses-of-the-future-possible-"},"modified":"2005-09-13T21:12:33","modified_gmt":"2005-09-14T02:12:33","slug":"gpgpu-and-oses-of-the-future-possible-resource-contention-problems","status":"publish","type":"post","link":"https:\/\/archive.blogs.harvard.edu\/hoanga\/2005\/09\/13\/gpgpu-and-oses-of-the-future-possible-resource-contention-problems\/","title":{"rendered":"GPGPU and OSes of the future&#8230; possible resource contention problems?"},"content":{"rendered":"<p><a name='a526'><\/a><\/p>\n<p>\n The <a href=\"http:\/\/en.wikipedia.org\/wiki\/Graphics_processing_unit\">GPU<\/a><br \/>\n has seen an amazing jump in power over the past few years.   It has moved<br \/>\n from just processing graphics to handling more general types of computations.<br \/>\n This general purpose computing on a GPU has been dubbed<br \/>\n <a href=\"http:\/\/en.wikipedia.org\/wiki\/GPGPU\">GPGPU<\/a> (General Purpose<br \/>\n GPU).  If the computation matches the computational model that the GPU<br \/>\n provides a tremendous speedup is possible.\n <\/p>\n<p>\n However, lately OSes have been learning to take advantage of using the<br \/>\n GPU for their own graphics operations.  Mac OS X tries to offload as much<br \/>\n of the fancy graphics effects on their desktop by giving it to the GPU<br \/>\n to process.  This is great as it leaves the main CPU free for the OS<br \/>\n and other applications to process more.\n <\/p>\n<p>\n But if GPGPU and OSes are both using the GPU very intensely this seems<br \/>\n to suggest that the GPU would become a contended resource in future<br \/>\n computing environments.   The rather straightforward answer to people<br \/>\n wanting to take advantage of GPU computing in this case is to buy another<br \/>\n GPU and use that as the co-processor and leave the main GPU for handling<br \/>\n the desktop.  But in a 1 GPU only environment I wonder if the resource<br \/>\n contention will become heavier in the future&#8230;\n <\/p>\n","protected":false},"excerpt":{"rendered":"<p>The GPU has seen an amazing jump in power over the past few years. It has moved from just processing graphics to handling more general types of computations. This general purpose computing on a GPU has been dubbed GPGPU (General Purpose GPU). If the computation matches the computational model that the GPU provides a tremendous [&hellip;]<\/p>\n","protected":false},"author":704,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-50","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"jetpack_featured_media_url":"","_links":{"self":[{"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/posts\/50","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/users\/704"}],"replies":[{"embeddable":true,"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/comments?post=50"}],"version-history":[{"count":0,"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/posts\/50\/revisions"}],"wp:attachment":[{"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/media?parent=50"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/categories?post=50"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/archive.blogs.harvard.edu\/hoanga\/wp-json\/wp\/v2\/tags?post=50"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}