{"id":5832,"date":"2023-09-20T14:18:39","date_gmt":"2023-09-20T21:18:39","guid":{"rendered":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/?p=5832"},"modified":"2023-12-19T11:20:03","modified_gmt":"2023-12-19T18:20:03","slug":"edgeengine-a-thermal-aware-optimization-framework-for-edge-inference","status":"publish","type":"post","link":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/2023\/09\/20\/edgeengine-a-thermal-aware-optimization-framework-for-edge-inference\/","title":{"rendered":"EdgeEngine: A Thermal-Aware Optimization Framework for Edge  Inference"},"content":{"rendered":"<p>Amirhossein Ahmadi, Hazem Abdelhafez, Karthik Pattabiraman and Matei Ripeanu, Proceedings of the <a href=\"https:\/\/acm-ieee-sec.org\/2023\/\">ACM\/IEEE International Symposium on Edge Computing (SEC), 2023<\/a>. (Acceptance Rate: 25%) [ <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/0bricew5ww4bdh4m3rgtt\/SEC_2023_Amir.pdf?rlkey=h6zlq0egefwsxy1cwra1s36yn&#038;dl=0\">PDF<\/a> | <a href=\"https:\/\/www.dropbox.com\/scl\/fi\/c72flshjckh7ryjbp20us\/SEC23-presentation.pdf?rlkey=loly09n2ulfs08gm1oqk71ivp&#038;dl=0\">Talk<\/a> ] (<a href=\"https:\/\/github.com\/DependableSystemsLab\/EdgeEngine\">Code<\/a>)<br \/>\n<!--more--><\/p>\n<p><strong>Abstract: <\/strong> Heterogeneous edge platforms enable the efficient execution of machine learning inference applications. These applications  often have a critical constraint (such as meeting a deadline) and an optimization goal (such as minimizing energy consumption). To navigate this space, existing optimization frameworks adjust the platform\u2019s frequency configuration for the CPU, the GPU and\/or the memory controller. However, existing optimization frameworks have two limitations. First, edge applications are frequently deployed in environments where they are exposed to ambient temperature variations. Second, a recent study has shown that temperature has a significant impact on edge platform characteristics. In this context, today\u2019s frequency optimization frameworks (which are thermal-oblivious) select frequency configurations that either violate the application\u2019s constraints, or are sub-optimal in terms of the optimization goal.<\/p>\n<p>To address these shortcomings, we propose EdgeEngine, a light-weight, thermal-aware optimization framework. EdgeEngine monitors the platform\u2019s temperature and uses reinforcement learning to adjust the frequency configuration of all underlying platform resources to meet the application\u2019s constraints. We find that EdgeEngine meets the application\u2019s constraint, and achieves 29% lower energy consumption (up to 2x) and 41% fewer violations compared to existing state-of-the-art thermal-oblivious optimization frameworks.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Amirhossein Ahmadi, Hazem Abdelhafez, Karthik Pattabiraman and Matei Ripeanu, Proceedings of the ACM\/IEEE International Symposium on Edge Computing (SEC), 2023. (Acceptance Rate: 25%) [ PDF | Talk ] (Code)<\/p>\n","protected":false},"author":10348,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5],"tags":[50,49,4,46,54],"class_list":["post-5832","post","type-post","status-publish","format-standard","hentry","category-publications","tag-50","tag-amir","tag-conference","tag-edge","tag-performance"],"_links":{"self":[{"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/posts\/5832","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/users\/10348"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/comments?post=5832"}],"version-history":[{"count":7,"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/posts\/5832\/revisions"}],"predecessor-version":[{"id":5889,"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/posts\/5832\/revisions\/5889"}],"wp:attachment":[{"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/media?parent=5832"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/categories?post=5832"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.ubc.ca\/dependablesystemslab\/wp-json\/wp\/v2\/tags?post=5832"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}