The difficulty of building large data centers in dense metro areas is pushing big cloud providers towards a different approach to scaling: multiple smaller data centers within tens of kilometers of each other, comprising a \u201cregion\u201d. We show that networking this small number of nearby sites with each other is a surprisingly challenging and multi-faceted problem. We draw out the operational goals and constraints of such networks, and highlight the design trade-offs involved using data from Microsoft Azure\u2019s regions.<\/p>\n
Our analysis of the design space shows that network topologies that achieve lower latency and allow greater flexibility in data center placement are, unfortunately, encumbered by their much greater cost and complexity. We thus present and demonstrate a novel optical-circuit-switched architecture, Iris, that lowers these cost and complexity barriers, making a richer topology design space more accessible to operators of regional networks. With Iris, topologies which, in comparison to a simple hub-and-spoke topology can increase the area in which a new DC can be placed by 2-5\u00d7, can be implemented at a cost within 1.1\u00d7 of the simple hub-and-spoke topology, and 7\u00d7 cheaper than a natural packet-switched network.<\/p>\n","protected":false},"excerpt":{"rendered":"
The difficulty of building large data centers in dense metro areas is pushing big cloud providers towards a different approach to scaling: multiple smaller data centers within tens of kilometers of each other, comprising a \u201cregion\u201d. We show that networking this small number of nearby sites with each other is a surprisingly challenging and multi-faceted […]<\/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":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13547],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-687240","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-systems-and-networking","msr-locale-en_us"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2020-8-1","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":"ACM","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:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2020\/08\/Iris_SIGCOMM2020.pdf","id":"687243","title":"iris_sigcomm2020","label_id":"243109","label":0}],"msr_related_uploader":"","msr_attachments":[{"id":687243,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2020\/08\/Iris_SIGCOMM2020.pdf"}],"msr-author-ordering":[{"type":"text","value":"Vojislav Dukic","user_id":0,"rest_url":false},{"type":"text","value":"Ginni Khanna","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Christos Gkantsidis","user_id":31424,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Christos Gkantsidis"},{"type":"user_nicename","value":"Thomas Karagiannis","user_id":34031,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Thomas Karagiannis"},{"type":"user_nicename","value":"Francesca Parmigiani","user_id":37727,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Francesca Parmigiani"},{"type":"text","value":"Ankit Singla","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Mark Filer","user_id":37715,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Mark Filer"},{"type":"guest","value":"jeff-cox","user_id":451530,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=jeff-cox"},{"type":"guest","value":"anna-ptasznik","user_id":610998,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=anna-ptasznik"},{"type":"guest","value":"nick-harland","user_id":611004,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=nick-harland"},{"type":"guest","value":"winston-saunders","user_id":611028,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=winston-saunders"},{"type":"user_nicename","value":"Christian Belady","user_id":31345,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Christian Belady"}],"msr_impact_theme":[],"msr_research_lab":[199561],"msr_event":[],"msr_group":[],"msr_project":[438654],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":438654,"post_title":"Project Iris","post_name":"iris","post_type":"msr-project","post_date":"2020-02-26 12:05:40","post_modified":"2023-03-24 03:18:07","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/iris\/","post_excerpt":"Project Iris explores novel designs of regional and Wide Area (WAN) cloud networks from the ground-up motivated by the rapid growth of cloud network traffic across data centers. Building upon recent advancements in optical technologies, we focus on investigating network architectures that are amenable to a distributed data center model and on designing the optical elements that can efficiently support and scale such architectures. These include both new hardware and data plane mechanisms such as…","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/438654"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/687240","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/687240\/revisions"}],"predecessor-version":[{"id":687246,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/687240\/revisions\/687246"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=687240"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=687240"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=687240"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=687240"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=687240"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=687240"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=687240"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=687240"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=687240"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=687240"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=687240"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=687240"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=687240"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=687240"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=687240"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=687240"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}