{"id":782347,"date":"2021-10-06T07:15:55","date_gmt":"2021-10-06T14:15:55","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-research-item&p=782347"},"modified":"2021-12-13T13:53:18","modified_gmt":"2021-12-13T21:53:18","slug":"kangaroo-caching-billions-of-tiny-objects-on-flash","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/kangaroo-caching-billions-of-tiny-objects-on-flash\/","title":{"rendered":"Kangaroo: Caching Billions of Tiny Objects on Flash"},"content":{"rendered":"

Many social-media and IoT services have very large working sets consisting of billions of tiny (~100 B) objects. Large, flash-based caches are important to serving these working sets at acceptable monetary cost. However, caching tiny objects on flash is challenging for two reasons: (i) SSDs can read\/write data only in multi-KB pages that are much larger than a single object, stressing the limited number of times flash can be written; and (ii) very few bits per cached object can be kept in DRAM without losing flash’s cost advantage. Unfortunately, existing flash-cache designs fall short of addressing these challenges: write-optimized designs require too much DRAM, and DRAM-optimized designs write flash too much.<\/p>\n

We present Kangaroo, a new flash-cache design that optimizes both DRAM usage and flash writes to maximize cache performance while minimizing cost. Kangaroo combines a large, set-associative cache with a small, log-structured cache. The set-associative cache requires minimal DRAM, while the log-structured cache minimizes Kangaroo’s flash writes. Experiments using traces from Facebook and Twitter show that Kangaroo achieves DRAM usage close to the best prior DRAM-optimized design, flash writes close to the best prior write-optimized design, and miss ratios better than both. Kangaroo’s design is Pareto-optimal across a range of allowed write rates, DRAM sizes, and flash sizes, reducing misses by 29% over the state of the art. These results are corroborated with a test deployment of Kangaroo in a production flash cache at Facebook.<\/p>\n","protected":false},"excerpt":{"rendered":"

Many social-media and IoT services have very large working sets consisting of billions of tiny (~100 B) objects. Large, flash-based caches are important to serving these working sets at acceptable monetary cost. However, caching tiny objects on flash is challenging for two reasons: (i) SSDs can read\/write data only in multi-KB pages that are much […]<\/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":[246574],"research-area":[13547],"msr-publication-type":[193726],"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-782347","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-highlight-award","msr-research-area-systems-and-networking","msr-locale-en_us"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2021-10-1","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":"SOSP 2021 Best Paper Award","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":"","msr_related_uploader":"","msr_attachments":[],"msr-author-ordering":[{"type":"text","value":"Sara McAllister","user_id":0,"rest_url":false},{"type":"text","value":"Benjamin Berg","user_id":0,"rest_url":false},{"type":"text","value":"Julian Tutuncu-Macias","user_id":0,"rest_url":false},{"type":"text","value":"Juncheng Yang","user_id":0,"rest_url":false},{"type":"text","value":"Sathya Gunasekar","user_id":0,"rest_url":false},{"type":"text","value":"Jimmy Lu","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Daniel Berger","user_id":38892,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Daniel Berger"},{"type":"text","value":"Nathan Beckmann","user_id":0,"rest_url":false},{"type":"text","value":"Gregory R. Ganger","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[144927],"msr_project":[],"publication":[],"video":[],"download":[],"msr_publication_type":"unpublished","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/782347"}],"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\/782347\/revisions"}],"predecessor-version":[{"id":782350,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/782347\/revisions\/782350"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=782347"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=782347"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=782347"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=782347"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=782347"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=782347"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=782347"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=782347"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=782347"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=782347"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=782347"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=782347"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=782347"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=782347"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=782347"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}