Data structures for non-volatile memories have to be designed such that they can be atomically modified using transactions. Existing atomicity methods require data to be copied in the critical path which significantly increases the\u00a0latency of transactions. These overheads are further amplified for transactions on byte-addressable persistent memories where often the byte ranges modified for data structure\u00a0updates are significantly smaller compared to the granularity at which data can be efficiently copied and logged.\u00a0We propose Kamino-Tx that provides a new way to perform transactional updates on non-volatile byte-addressable\u00a0memories (NVM) without requiring any copying of data in\u00a0the critical path. Kamino-Tx maintains an additional copy of\u00a0data off the critical path to achieve atomicity. But in doing\u00a0so Kamino-Tx has to overcome two important challenges of\u00a0safety and minimizing NVM storage overhead. We propose\u00a0a more dynamic approach to maintaining the additional copy\u00a0of data to reduce storage overheads. To further mitigate the\u00a0storage overhead of using Kamino-Tx in a replicated setting,\u00a0we develop Kamino-Tx-Chain, a variant of Chain Replication where replicas perform in-place updates and do not\u00a0maintain data copies locally; replicas in Kamino-Tx-Chain\u00a0leverage other replicas as copies to roll back or forward for\u00a0atomicity. Our results show that using Kamino-Tx increases\u00a0throughput by up to 9.5x for unreplicated systems and up to\u00a02.2x for replicated settings.<\/p>\n","protected":false},"excerpt":{"rendered":"
Data structures for non-volatile memories have to be designed such that they can be atomically modified using transactions. Existing atomicity methods require data to be copied in the critical path which significantly increases the\u00a0latency of transactions. These overheads are further amplified for transactions on byte-addressable persistent memories where often the byte ranges modified for data […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"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-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-356759","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-systems-and-networking","msr-locale-en_us"],"msr_publishername":"","msr_edition":"European Conference on Computer Systems (EuroSys'17)","msr_affiliation":"","msr_published_date":"2017-04-23","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":"374654","msr_publicationurl":"http:\/\/eurosys2017.org\/","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"paper","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2017\/01\/paper-2.pdf","id":374654,"label_id":0},{"type":"url","title":"http:\/\/eurosys2017.org\/","viewUrl":false,"id":false,"label_id":0}],"msr_related_uploader":"","msr_attachments":[{"id":0,"url":"http:\/\/eurosys2017.org\/"}],"msr-author-ordering":[{"type":"text","value":"Amir Saman Memaripour","user_id":0,"rest_url":false},{"type":"user_nicename","value":"anbadam","user_id":31008,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=anbadam"},{"type":"user_nicename","value":"amar","user_id":30975,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=amar"},{"type":"text","value":"Ram Alagappan","user_id":0,"rest_url":false},{"type":"text","value":"Yanqi Zhou","user_id":0,"rest_url":false},{"type":"user_nicename","value":"kstrauss","user_id":32587,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=kstrauss"},{"type":"text","value":"Steven Swanson","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[144927],"msr_project":[171456],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/356759"}],"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":3,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/356759\/revisions"}],"predecessor-version":[{"id":531393,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/356759\/revisions\/531393"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=356759"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=356759"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=356759"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=356759"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=356759"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=356759"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=356759"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=356759"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=356759"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=356759"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=356759"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=356759"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=356759"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=356759"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=356759"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}