{"id":268836,"date":"2016-06-22T07:15:32","date_gmt":"2016-06-22T14:15:32","guid":{"rendered":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/?post_type=msr-research-item&#038;p=268836"},"modified":"2018-10-16T21:02:42","modified_gmt":"2018-10-17T04:02:42","slug":"fast-and-cautious-leveraging-multi-path-diversity-for-transport-loss-recovery-in-data-centers","status":"publish","type":"msr-research-item","link":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/publication\/fast-and-cautious-leveraging-multi-path-diversity-for-transport-loss-recovery-in-data-centers\/","title":{"rendered":"Fast and Cautious: Leveraging Multi-path Diversity for Transport Loss Recovery in Data Centers"},"content":{"rendered":"<p>To achieve low TCP flow completion time (FCT) in data center networks (DCNs), it is critical and challenging to rapidly recover loss without adding extra congestion. Therefore, in this paper we propose a novel loss recovery approach FUSO that exploits multi-path diversity in DCN for transport loss recovery. In FUSO, when a multi-path transport sender suspects loss on one subflow, recovery packets are immediately sent over another sub-flow that is not or less lossy <em>and<\/em> has spare congestion window slots. FUSO is <em>fast<\/em> in that it does not need to wait for timeout on the lossy sub-flow, and it is <em>cautious<\/em> in that it does not violate congestion control algorithm. Testbed experiments and simulations show that FUSO decreases the latency-sensitive flows\u2019 99<em>th<\/em> percentile FCT by up to ~82.3% in a 1Gbps testbed, and up to ~87.9% in a 10Gpbs large-scale simulated network.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>To achieve low TCP flow completion time (FCT) in data center networks (DCNs), it is critical and challenging to rapidly recover loss without adding extra congestion. Therefore, in this paper we propose a novel loss recovery approach FUSO that exploits multi-path diversity in DCN for transport loss recovery. In FUSO, when a multi-path transport sender [&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":"USENIX","msr_publisher_other":"","msr_booktitle":"","msr_chapter":"","msr_edition":"Annual Technical Conference","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":"Annual Technical Conference","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":"2016-06-22","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":0,"footnotes":""},"msr-research-highlight":[],"research-area":[13547],"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-268836","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-systems-and-networking","msr-locale-en_us"],"msr_publishername":"USENIX","msr_edition":"Annual Technical Conference","msr_affiliation":"","msr_published_date":"2016-06-22","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":"268839","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"atc16_paper_chen","viewUrl":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-content\/uploads\/2016\/08\/atc16_paper_chen.pdf","id":268839,"label_id":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":[],"msr-author-ordering":[{"type":"text","value":"Yuanwei Lu","user_id":0,"rest_url":false},{"type":"text","value":"Yuan Meng","user_id":0,"rest_url":false},{"type":"text","value":"Dan Pei","user_id":0,"rest_url":false},{"type":"user_nicename","value":"pengc","user_id":33225,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=pengc"},{"type":"text","value":"Larry Luo","user_id":0,"rest_url":false},{"type":"user_nicename","value":"yqx","user_id":35049,"rest_url":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=yqx"},{"type":"text","value":"Xiaoliang Wang","user_id":0,"rest_url":false},{"type":"text","value":"Youjian Zhao","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[199560],"msr_event":[],"msr_group":[144801],"msr_project":[268788],"publication":[],"video":[],"msr-tool":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":268788,"post_title":"FUSO: A Fast Multi-path Transport Loss Recovery Scheme for Data Centers","post_name":"fuso-fast-multi-path-transport-loss-recovery-scheme-data-centers-2","post_type":"msr-project","post_date":"2016-08-01 04:04:23","post_modified":"2016-08-11 22:44:19","post_status":"publish","permalink":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/project\/fuso-fast-multi-path-transport-loss-recovery-scheme-data-centers-2\/","post_excerpt":"Packet loss in data centers, caused by both congestion and failures, greatly hurts the performance of the transport layer, leading to a long tail of flow completion time. FUSO is a fast multi-path transport loss recovery scheme for data centers, to help maintaining a consistent low flow completion time when facing packet losses. FUSO leverages the multi-path diversity in data centers to accelerate the loss recovery, attempting to be both \"fast\" and \"cautious\". Specifically, when\u00a0a&hellip;","_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/268788"}]}}]},"_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/268836","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":1,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/268836\/revisions"}],"predecessor-version":[{"id":422661,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/268836\/revisions\/422661"}],"wp:attachment":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/media?parent=268836"}],"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=268836"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=268836"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=268836"},{"taxonomy":"msr-publisher","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-publisher?post=268836"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=268836"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=268836"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=268836"},{"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=268836"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=268836"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=268836"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=268836"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=268836"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}