{"id":852000,"date":"2022-06-13T12:07:35","date_gmt":"2022-06-13T19:07:35","guid":{"rendered":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/"},"modified":"2025-09-02T07:24:20","modified_gmt":"2025-09-02T14:24:20","slug":"measuring-the-carbon-intensity-of-ai-in-cloud-instances","status":"publish","type":"msr-research-item","link":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/publication\/measuring-the-carbon-intensity-of-ai-in-cloud-instances\/","title":{"rendered":"Measuring the Carbon Intensity of AI in Cloud Instances"},"content":{"rendered":"<p><span dir=\"ltr\" role=\"presentation\">The advent of cloud computing has provided people around the world with unprecedented access to computational power and <\/span><span dir=\"ltr\" role=\"presentation\">enabled rapid growth in technologies such as machine learning, the computational demands of which incur a high energy cost and a <\/span><span dir=\"ltr\" role=\"presentation\">commensurate carbon footprint. As a result, recent scholarship has called for better estimates of the greenhouse gas impact of AI: data<\/span><br role=\"presentation\" \/><span dir=\"ltr\" role=\"presentation\">scientists today do not have easy or reliable access to measurements of this information, which precludes development of actionable <\/span><span dir=\"ltr\" role=\"presentation\">tactics. We argue that cloud providers presenting information about software carbon intensity to users is a fundamental stepping <\/span><span dir=\"ltr\" role=\"presentation\">stone towards minimizing emissions. <\/span><span dir=\"ltr\" role=\"presentation\">In this paper, we provide a framework for measuring software carbon intensity, and propose to measure operational carbon <\/span><span dir=\"ltr\" role=\"presentation\">emissions by using location-based and time-specific marginal emissions data per energy unit. We provide measurements of operational <\/span><span dir=\"ltr\" role=\"presentation\">software carbon intensity for a set of modern models covering natural language processing and computer vision applications, and a <\/span><span dir=\"ltr\" role=\"presentation\">wide range of model sizes, including pretraining of a 6.1 billion parameter language model. We then evaluate a suite of approaches for <\/span><span dir=\"ltr\" role=\"presentation\">reducing emissions on the Microsoft Azure cloud compute platform: using cloud instances in different geographic regions, using cloud <\/span><span dir=\"ltr\" role=\"presentation\">instances at different times of day, and dynamically pausing cloud instances when the marginal carbon intensity is above a certain <\/span><span dir=\"ltr\" role=\"presentation\">threshold. We confirm previous results that the geographic region of the data center plays a significant role in the carbon intensity for <\/span><span dir=\"ltr\" role=\"presentation\">a given cloud instance, and find that choosing an appropriate region can have the largest operational emissions reduction impact. We <\/span><span dir=\"ltr\" role=\"presentation\">also present new results showing that the time of day has meaningful impact on operational software carbon intensity. Finally, we <\/span><span dir=\"ltr\" role=\"presentation\">conclude with recommendations for how machine learning practitioners can use software carbon intensity information to reduce <\/span><span dir=\"ltr\" role=\"presentation\">environmental impact.<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>The advent of cloud computing has provided people around the world with unprecedented access to computational power and enabled rapid growth in technologies such as machine learning, the computational demands of which incur a high energy cost and a commensurate carbon footprint. As a result, recent scholarship has called for better estimates of the greenhouse [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-author-ordering":null,"msr_publishername":"","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"","msr_editors":"","msr_how_published":"","msr_isbn":"","msr_issue":"","msr_journal":"","msr_number":"","msr_organization":"","msr_pages_string":"","msr_page_range_start":"","msr_page_range_end":"","msr_series":"","msr_volume":"","msr_copyright":"","msr_conference_name":"ACM Conference on Fairness, Accountability, and Transparency","msr_doi":"","msr_arxiv_id":"","msr_s2_paper_id":"","msr_mag_id":"","msr_pubmed_id":"","msr_other_authors":"","msr_other_contributors":"","msr_speaker":"","msr_award":"","msr_affiliation":"","msr_institution":"","msr_host":"","msr_version":"","msr_duration":"","msr_original_fields_of_study":"","msr_release_tracker_id":"","msr_s2_match_type":"","msr_citation_count_updated":"","msr_published_date":"2022-6-1","msr_highlight_text":"","msr_notes":"","msr_longbiography":"","msr_publicationurl":"","msr_external_url":"","msr_secondary_video_url":"","msr_conference_url":"","msr_journal_url":"","msr_s2_pdf_url":"","msr_year":0,"msr_citation_count":0,"msr_influential_citations":0,"msr_reference_count":0,"msr_s2_match_confidence":0,"msr_microsoftintellectualproperty":true,"msr_s2_open_access":false,"msr_s2_author_ids":[],"msr_pub_ids":[],"msr_hide_image_in_river":null,"footnotes":""},"msr-research-highlight":[],"research-area":[13556,198583],"msr-publication-type":[193716],"msr-publisher":[],"msr-focus-area":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-852000","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-ecology-environment","msr-locale-en_us"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2022-6-1","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","viewUrl":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-content\/uploads\/2022\/06\/CO2_azure_paper.pdf","id":"852006","title":"co2_azure_paper","label_id":"243109","label":0}],"msr_related_uploader":"","msr_citation_count":0,"msr_citation_count_updated":"","msr_s2_paper_id":"","msr_influential_citations":0,"msr_reference_count":0,"msr_arxiv_id":"","msr_s2_author_ids":[],"msr_s2_open_access":false,"msr_s2_pdf_url":null,"msr_attachments":[{"id":852006,"url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-content\/uploads\/2022\/06\/CO2_azure_paper.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"Remi Tachet des Combes","user_id":37086,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Remi Tachet des Combes"},{"type":"user_nicename","value":"Erika Odmark","user_id":41737,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Erika Odmark"},{"type":"guest","value":"will-buchanan","user_id":830671,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=will-buchanan"}],"msr_impact_theme":[],"msr_research_lab":[437514],"msr_event":[848407],"msr_group":[896463,1148823],"msr_project":[804847],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":804847,"post_title":"Reducing AI's Carbon Footprint","post_name":"reducing-ais-carbon-footprint","post_type":"msr-project","post_date":"2022-05-24 08:56:55","post_modified":"2024-01-16 11:11:59","post_status":"publish","permalink":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/project\/reducing-ais-carbon-footprint\/","post_excerpt":"This project develops techniques that enable AI to use computing infrastructure more efficiently. The goals are to maintain predictive accuracy while reducing carbon emissions, whether embodied in manufactured hardware, or produced from electricity usage when green energy is not available.","_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/804847"}]}}]},"_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/852000","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":4,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/852000\/revisions"}],"predecessor-version":[{"id":1148957,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/852000\/revisions\/1148957"}],"wp:attachment":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/media?parent=852000"}],"wp:term":[{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=852000"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=852000"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=852000"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=852000"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=852000"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=852000"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=852000"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=852000"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=852000"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=852000"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=852000"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=852000"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}