{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,3]],"date-time":"2025-11-03T23:03:58Z","timestamp":1762211038203,"version":"3.41.0"},"reference-count":75,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2022,8,24]],"date-time":"2022-08-24T00:00:00Z","timestamp":1661299200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"NDSEG Fellowship"},{"name":"Google Research Scholar Award","award":["NSF-CMMI-1938909, NSF-CSR-1763701"],"award-info":[{"award-number":["NSF-CMMI-1938909, NSF-CSR-1763701"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Storage"],"published-print":{"date-parts":[[2022,8,31]]},"abstract":"<jats:p>\n            Many social-media and IoT services have very large working sets consisting of billions of tiny (\u2248100\u00a0B) 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:\n            <jats:italic>(i)<\/jats:italic>\n            \u00a0SSDs can read\/write data only in multi-KB \u201cpages\u201d that are much larger than a single object, stressing the limited number of times flash can be written; and\n            <jats:italic>(ii)<\/jats:italic>\n            \u00a0very few bits per cached object can be kept in DRAM without losing flash\u2019s cost advantage. Unfortunately, existing flash-cache designs fall short of addressing these challenges: write-optimized designs require too much DRAM, and DRAM-optimized designs require too many flash writes.\n          <\/jats:p>\n          <jats:p>\n            We present\n            <jats:sc>Kangaroo<\/jats:sc>\n            , 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\u2019s flash writes. Experiments using traces from Meta 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\u2019s 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 by analytical models presented herein and with a test deployment of Kangaroo in a production flash cache at Meta.\n          <\/jats:p>","DOI":"10.1145\/3542928","type":"journal-article","created":{"date-parts":[[2022,6,13]],"date-time":"2022-06-13T12:22:03Z","timestamp":1655122923000},"page":"1-33","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":4,"title":["Kangaroo: Theory and Practice of Caching Billions of Tiny Objects on Flash"],"prefix":"10.1145","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5253-7094","authenticated-orcid":false,"given":"Sara","family":"McAllister","sequence":"first","affiliation":[{"name":"Carnegie Mellon University, Pittsburgh, PA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4147-6860","authenticated-orcid":false,"given":"Benjamin","family":"Berg","sequence":"additional","affiliation":[{"name":"Carnegie Mellon University, Pittsburgh, PA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4286-7721","authenticated-orcid":false,"given":"Julian","family":"Tutuncu-Macias","sequence":"additional","affiliation":[{"name":"Goldman Sachs, New York City, , NY"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0412-1139","authenticated-orcid":false,"given":"Juncheng","family":"Yang","sequence":"additional","affiliation":[{"name":"Carnegie Mellon University, Pittsburgh, PA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8885-0098","authenticated-orcid":false,"given":"Sathya","family":"Gunasekar","sequence":"additional","affiliation":[{"name":"Meta, Menlo Park"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0560-9160","authenticated-orcid":false,"given":"Jimmy","family":"Lu","sequence":"additional","affiliation":[{"name":"Meta, Menlo Park"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3911-1512","authenticated-orcid":false,"given":"Daniel S.","family":"Berger","sequence":"additional","affiliation":[{"name":"Microsoft Research and University of Washington, Redmond, Washington"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6301-714X","authenticated-orcid":false,"given":"Nathan","family":"Beckmann","sequence":"additional","affiliation":[{"name":"Carnegie Mellon University, Pittsburgh, PA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3065-7316","authenticated-orcid":false,"given":"Gregory R.","family":"Ganger","sequence":"additional","affiliation":[{"name":"Carnegie Mellon University, Pittsburgh, PA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2022,8,24]]},"reference":[{"key":"e_1_3_2_2_2","unstructured":"https:\/\/aws.amazon.com\/dynamodb\/features\/. 5\/5\/21 Amazon DynamoDB"},{"key":"e_1_3_2_3_2","unstructured":"https:\/\/trafficserver.apache.org Apache Traffic Server"},{"key":"e_1_3_2_4_2","unstructured":"https:\/\/azure.microsoft.com\/en-us\/services\/cache\/#what-you-can-build. 5\/5\/21 Azure Cache for Redis"},{"key":"e_1_3_2_5_2","unstructured":"https:\/\/saasscout.com\/statistics\/big-data-statistics\/. 5\/6\/21 Big Data Statistics Growth and Facts 2020"},{"key":"e_1_3_2_6_2","unstructured":"https:\/\/github.com\/twitter\/fatcache Fatcache"},{"key":"e_1_3_2_7_2","unstructured":"https:\/\/github.com\/google\/leveldb LevelDB"},{"key":"e_1_3_2_8_2","unstructured":"https:\/\/docs.redislabs.com\/latest\/rs\/concepts\/memory-architecture\/redis-flash\/ Redis on Flash"},{"key":"e_1_3_2_9_2","unstructured":"http:\/\/rocksdb.org RocksDB"},{"key":"e_1_3_2_10_2","unstructured":"investor.fb.com 2020 Facebook reports first quarter 2020 results"},{"key":"e_1_3_2_11_2","unstructured":"investor.twitterinc.com 2021 Twitter first quarter 2021 results"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1145\/321623.321632"},{"key":"e_1_3_2_13_2","volume-title":"USENIX NSDI","author":"Beckmann Nathan","year":"2018","unstructured":"Nathan Beckmann, Haoxian Chen, and Asaf Cidon. 2018. LHD: Improving hit rate by maximizing hit density. In USENIX NSDI."},{"key":"e_1_3_2_14_2","volume-title":"IEEE HPCA","author":"Beckmann Nathan","year":"2015","unstructured":"Nathan Beckmann and Daniel Sanchez. 2015. Talus: A simple way to remove cliffs in cache performance. In IEEE HPCA."},{"key":"e_1_3_2_15_2","volume-title":"IEEE HPCA","author":"Beckmann Nathan","year":"2015","unstructured":"Nathan Beckmann, Po-An Tsai, and Daniel Sanchez. 2015. Scaling distributed cache hierarchies through computation and data co-scheduling. In IEEE HPCA."},{"key":"e_1_3_2_16_2","volume-title":"ACM SPAA","author":"Bender Michael A.","year":"2019","unstructured":"Michael A. Bender, Alex Conway, Mart\u00edn Farach-Colton, William Jannen, Yizheng Jiao, Rob Johnson, Eric Knorr, Sara McAllister, Nirjhar Mukherjee, Prashant Pandey, Donald E. Porter, Jun Yuan, and Yang Zhan. 2019. Small refinements to the DAM can have big consequences for data-structure design. In ACM SPAA."},{"key":"e_1_3_2_17_2","volume-title":"USENIX OSDI","author":"Berg Benjamin","year":"2020","unstructured":"Benjamin Berg, Daniel S. Berger, Sara McAllister, Isaac Grosof, Sathya Gunasekar, Jimmy Lu, Michael Uhlar, Jim Carrig, Nathan Beckmann, Mor Harchol-Balter, and Gregory G. Ganger. 2020. The CacheLib caching engine: Design and experiences at scale. In USENIX OSDI."},{"key":"e_1_3_2_18_2","volume-title":"ACM SIGMETRICS","author":"Berger Daniel S.","year":"2018","unstructured":"Daniel S. Berger, Nathan Beckmann, and Mor Harchol-Balter. 2018. Practical bounds on optimal caching with variable object sizes. In ACM SIGMETRICS."},{"key":"e_1_3_2_19_2","volume-title":"USENIX OSDI","author":"Berger Daniel S.","year":"2018","unstructured":"Daniel S. Berger, Benjamin Berg, Timothy Zhu, Siddhartha Sen, and Mor Harchol-Balter. 2018. Robinhood: Tail latency aware caching\u2013dynamic reallocation from cache-rich to cache-poor. In USENIX OSDI."},{"key":"e_1_3_2_20_2","volume-title":"USENIX NSDI","author":"Berger Daniel S.","year":"2017","unstructured":"Daniel S. Berger, Ramesh K. Sitaraman, and Mor Harchol-Balter. 2017. AdaptSize: Orchestrating the hot object memory cache in a content delivery network. In USENIX NSDI."},{"key":"e_1_3_2_21_2","volume-title":"USENIX ATC.","author":"Blankstein Aaron","year":"2017","unstructured":"Aaron Blankstein, Siddhartha Sen, and Michael J. Freedman. 2017. Hyperbolic caching: Flexible caching for web applications. In USENIX ATC."},{"key":"e_1_3_2_22_2","article-title":"Application Data Caching Using SSDs","author":"Blog Netflix Technology","year":"2016","unstructured":"Netflix Technology Blog. 2016. Application Data Caching Using SSDs. https:\/\/netflixtechblog.com\/application-data-caching-using-ssds-5bf25df851ef.","journal-title":"https:\/\/netflixtechblog.com\/application-data-caching-using-ssds-5bf25df851ef"},{"key":"e_1_3_2_23_2","article-title":"Evolution of Application Data Caching: From RAM to SSD","author":"Blog Netflix Technology","year":"2018","unstructured":"Netflix Technology Blog. 2018. Evolution of Application Data Caching: From RAM to SSD. https:\/\/netflixtechblog.com\/evolution-of-application-data-caching-from-ram-to-ssd-a33d6fa7a690.","journal-title":"https:\/\/netflixtechblog.com\/evolution-of-application-data-caching-from-ram-to-ssd-a33d6fa7a690"},{"key":"e_1_3_2_24_2","volume-title":"USENIX FAST","author":"Boboila Simona","year":"2010","unstructured":"Simona Boboila and Peter Desnoyers. 2010. Write endurance in flash drives: Measurements and analysis. In USENIX FAST."},{"key":"e_1_3_2_25_2","volume-title":"USENIX ATC","author":"Bronson Nathan","year":"2013","unstructured":"Nathan Bronson, Zach Amsden, George Cabrera, Prasad Chakka, Peter Dimov, Hui Ding, Jack Ferris, Anthony Giardullo, Sachin Kulkarni, Harry Li, Mark Marchukov, Dmitri Petrov, Lovro Puzar, Yee Jiun Song, and Venkat Venkataramani. 2013. TAO: Facebook\u2019s distributed data store for the social graph. In USENIX ATC."},{"key":"e_1_3_2_26_2","article-title":"FASTER: An embedded concurrent key-value store for state management","author":"Chandramouli Badrish","year":"2018","unstructured":"Badrish Chandramouli, Guna Prasaad, Donald Kossmann, Justin Levandoski, James Hunter, and Mike Barnett. 2018. FASTER: An embedded concurrent key-value store for state management. VLDB (2018).","journal-title":"VLDB"},{"key":"e_1_3_2_27_2","doi-asserted-by":"publisher","DOI":"10.1145\/1365815.1365816"},{"key":"e_1_3_2_28_2","volume-title":"USENIX HotStorage","author":"Chatzieleftheriou Andromachi","year":"2020","unstructured":"Andromachi Chatzieleftheriou, Ioan Stefanovici, Dushyanth Narayanan, Benn Thomsen, and Antony Rowstron. 2020. Could cloud storage be disrupted in the next decade? In USENIX HotStorage."},{"key":"e_1_3_2_29_2","volume-title":"USENIX NSDI","author":"Cidon Asaf","year":"2016","unstructured":"Asaf Cidon, Assaf Eisenman, Mohammad Alizadeh, and Sachin Katti. 2016. Cliffhanger: Scaling performance cliffs in web memory caches. In USENIX NSDI."},{"key":"e_1_3_2_30_2","volume-title":"USENIX ATC","author":"Cidon Asaf","year":"2017","unstructured":"Asaf Cidon, Daniel Rushton, Stephen M. Rumble, and Ryan Stutsman. 2017. Memshare: A dynamic multi-tenant key-value cache. In USENIX ATC."},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.5555\/574894"},{"key":"e_1_3_2_32_2","volume-title":"ACM SIGMETRICS.","author":"Dan Asit","year":"1990","unstructured":"Asit Dan and Don Towsley. 1990. An approximate analysis of the LRU and FIFO buffer replacement schemes. In ACM SIGMETRICS."},{"key":"e_1_3_2_33_2","first-page":"1","volume-title":"IEEE International Conference on Consumer Electronics","author":"Davis Gary","year":"2018","unstructured":"Gary Davis. 2018. 2020: Life with 50 billion connected devices. In IEEE International Conference on Consumer Electronics. 1\u20131."},{"key":"e_1_3_2_34_2","volume-title":"ACM SIGMOD","author":"Debnath Biplob","year":"2011","unstructured":"Biplob Debnath, Sudipta Sengupta, and Jin Li. 2011. SkimpyStash: RAM space skimpy key-value store on flash-based storage. In ACM SIGMOD."},{"key":"e_1_3_2_35_2","volume-title":"USENIX NSDI","author":"Eisenman Assaf","year":"2019","unstructured":"Assaf Eisenman, Asaf Cidon, Evgenya Pergament, Or Haimovich, Ryan Stutsman, Mohammad Alizadeh, and Sachin Katti. 2019. Flashield: A hybrid key-value cache that controls flash write amplification. In USENIX NSDI."},{"key":"e_1_3_2_36_2","volume-title":"USENIX NSDI","author":"Fan Bin","year":"2013","unstructured":"Bin Fan, David G. Andersen, and Michael Kaminsky. 2013. MemC3: Compact and concurrent memcache with dumber caching and smarter hashing. In USENIX NSDI."},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","DOI":"10.5555\/1012889.1012894"},{"key":"e_1_3_2_38_2","article-title":"Microsoft","author":"Freiling Peter","unstructured":"Peter Freiling and Badrish Chandramouli.[n. d.]. Microsoft. Personal communication.","journal-title":"Personal communication"},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.1145\/2318857.2254810"},{"key":"e_1_3_2_40_2","article-title":"McDipper: A Key-value Cache for Flash Storage","author":"Gartrell Alex","unstructured":"Alex Gartrell, Mohan Srinivasan, Bryan Alger, and Kumar Sundararajan.[n. d.]. McDipper: A Key-value Cache for Flash Storage. https:\/\/www.facebook.com\/notes\/facebook-engineering\/mcdipper-a-key-value-cache-for-flash-storage\/10151347090423920\/.","journal-title":"https:\/\/www.facebook.com\/notes\/facebook-engineering\/mcdipper-a-key-value-cache-for-flash-storage\/10151347090423920\/"},{"key":"e_1_3_2_41_2","volume-title":"ACM EuroSys","author":"He Jun","year":"2017","unstructured":"Jun He, Sudarsun Kannan, Andrea C. Arpaci-Dusseau, and Remzi H. Arpaci-Dusseau. 2017. The unwritten contract of solid state drives. In ACM EuroSys."},{"key":"e_1_3_2_42_2","volume-title":"ISCA-37","author":"Jaleel Aamer","year":"2010","unstructured":"Aamer Jaleel, Kevin Theobald, Simon Steely Jr., and Joel Emer. 2010. High performance cache replacement using re-reference interval prediction. In ISCA-37."},{"key":"e_1_3_2_43_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0167-6377(99)00037-1"},{"key":"e_1_3_2_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/12.286300"},{"key":"e_1_3_2_45_2","unstructured":"Changman Lee Dongho Sim Jooyoung Hwang and Sangyeun Cho. 2015. F2FS: A new file system for flash storage."},{"key":"e_1_3_2_46_2","volume-title":"ACM SoCC","author":"Li Conglong","year":"2017","unstructured":"Conglong Li, David G. Andersen, Qiang Fu, Sameh Elnikety, and Yuxiong He. 2017. Workload analysis and caching strategies for search advertising systems. In ACM SoCC."},{"key":"e_1_3_2_47_2","first-page":"1875","volume-title":"WWW.","author":"Li Conglong","year":"2018","unstructured":"Conglong Li, David G. Andersen, Qiang Fu, Sameh Elnikety, and Yuxiong He. 2018. Better caching in search advertising systems with rapid refresh predictions. In WWW.1875\u20131884."},{"key":"e_1_3_2_48_2","doi-asserted-by":"publisher","DOI":"10.1145\/3094785"},{"key":"e_1_3_2_49_2","volume-title":"ACM SOSP","author":"Lim Hyeontaek","year":"2011","unstructured":"Hyeontaek Lim, Bin Fan, David G. Andersen, and Michael Kaminsky. 2011. SILT: A memory-efficient, high-performance key-value store. In ACM SOSP."},{"key":"e_1_3_2_50_2","volume-title":"USENIX FAST","author":"Lu Lanyue","year":"2016","unstructured":"Lanyue Lu, Thanumalayan Sankaranarayana Pillai, Andrea C. Arpaci-Dusseau, and Remzi H. Arpaci-Dusseau. 2016. WiscKey: Separating keys from values in SSD-conscious storage. In USENIX FAST."},{"key":"e_1_3_2_51_2","volume-title":"ACM SOSP","author":"Mammarella Mike","year":"2009","unstructured":"Mike Mammarella, Shant Hovsepian, and Eddie Kohler. 2009. Modular data storage with Anvil. In ACM SOSP."},{"key":"e_1_3_2_52_2","volume-title":"ASPLOS","author":"Mukkara Anurag","year":"2016","unstructured":"Anurag Mukkara, Nathan Beckmann, and Daniel Sanchez. 2016. Whirlpool: Improving dynamic cache management with static data classification. In ASPLOS."},{"key":"e_1_3_2_53_2","volume-title":"USENIX OSDI","author":"O\u2019Toole James","year":"1994","unstructured":"James O\u2019Toole and Liuba Shrira. 1994. Opportunistic log: Efficient installation reads in a reliable storage server. In USENIX OSDI."},{"key":"e_1_3_2_54_2","doi-asserted-by":"publisher","DOI":"10.5555\/77493"},{"key":"e_1_3_2_55_2","doi-asserted-by":"publisher","DOI":"10.1109\/TNET.2015.2394482"},{"key":"e_1_3_2_56_2","article-title":"Twitter\u2019s doubling of character count from 140 to 280 had little impact on length of tweets","author":"Perez Sara","year":"2018","unstructured":"Sara Perez. 2018. Twitter\u2019s doubling of character count from 140 to 280 had little impact on length of tweets. Tech Crunch (2018). Available at https:\/\/techcrunch.com\/2018\/10\/30\/twitters-doubling-of-character-count-from-140-to-280-had-little-impact-on-length-of-tweets\/. 5\/4\/2021.","journal-title":"Tech Crunch"},{"key":"e_1_3_2_57_2","volume-title":"ACM SOSP","author":"Raju Pandian","year":"2017","unstructured":"Pandian Raju, Rohan Kadekodi, Vijay Chidambaram, and Ittai Abraham. 2017. PebblesDB: Building key-value stores using fragmented log-structured merge trees. In ACM SOSP."},{"key":"e_1_3_2_58_2","volume-title":"ACM SOSP","author":"Rosenblum Mendel","year":"1991","unstructured":"Mendel Rosenblum and John K. Ousterhout. 1991. The design and implementation of a log-structured file system. In ACM SOSP."},{"key":"e_1_3_2_59_2","volume-title":"IEEE INFOCOM","author":"Rosensweig Elisha J.","year":"2010","unstructured":"Elisha J. Rosensweig, Jim Kurose, and Don Towsley. 2010. Approximate models for general cache networks. In IEEE INFOCOM."},{"key":"e_1_3_2_60_2","volume-title":"ACM SIGMETRICS.","author":"Sen Rathijit","year":"2013","unstructured":"Rathijit Sen and David A. Wood. 2013. Reuse-based online models for caches. In ACM SIGMETRICS."},{"key":"e_1_3_2_61_2","volume-title":"USENIX HotStorage","author":"Shen Zhaoyan","year":"2016","unstructured":"Zhaoyan Shen, Feng Chen, Yichen Jia, and Zili Shao. 2016. Optimizing flash-based key-value cache systems. In USENIX HotStorage."},{"key":"e_1_3_2_62_2","doi-asserted-by":"publisher","DOI":"10.1145\/3203410"},{"key":"e_1_3_2_63_2","volume-title":"USENIX NSDI","author":"Song Zhenyu","year":"2020","unstructured":"Zhenyu Song, Daniel S. Berger, Kai Li, and Wyatt Lloyd. 2020. Learning relaxed Belady for content distribution network caching. In USENIX NSDI."},{"issue":"1","key":"e_1_3_2_64_2","article-title":"Anywhere the eye can see, it\u2019s likely to see an ad","volume":"15","author":"Story Louise","year":"2007","unstructured":"Louise Story. 2007. Anywhere the eye can see, it\u2019s likely to see an ad. The New York Times 15, 1 (2007). Available at https:\/\/www.nytimes.com\/2007\/01\/15\/business\/media\/15everywhere.html. 9\/6\/2020.","journal-title":"The New York Times"},{"key":"e_1_3_2_65_2","volume-title":"USENIX OSDI.","author":"Tang Chunqiang","year":"2020","unstructured":"Chunqiang Tang, Kenny Yu, Kaushik Veeraraghavan, Jonathan Kaldor, Scott Michelson, Thawan Kooburat, Aravind Anbudurai, Matthew Clark, Kabir Gogia, Long Cheng, Ben Christensen, Alex Gartrell, Maxim Khutornenko, Sachin Kulkarni, Marcin Pawlowski, Tuomas Pelkonen, Andre Rodrigues, Rounak Tibrewal, Vaishnavi Venkatesan, and Peter Zhang. 2020. Twine: A unified cluster management system for shared infrastructure. In USENIX OSDI."},{"key":"e_1_3_2_66_2","volume-title":"USENIX FAST","author":"Tang Linpeng","year":"2015","unstructured":"Linpeng Tang, Qi Huang, Wyatt Lloyd, Sanjeev Kumar, and Kai Li. 2015. RIPQ: Advanced photo caching on flash for Facebook. In USENIX FAST."},{"key":"e_1_3_2_67_2","article-title":"The anatomy of the Facebook social graph","author":"Ugander Johan","year":"2011","unstructured":"Johan Ugander, Brian Karrer, Lars Backstrom, and Cameron Marlow. 2011. The anatomy of the Facebook social graph. arXiv preprint arXiv:1111.4503 (2011).","journal-title":"arXiv preprint arXiv:1111.4503"},{"key":"e_1_3_2_68_2","volume-title":"The Varnish Book for Varnish 4.0","author":"Vel\u00e1zquez Francisco","year":"2016","unstructured":"Francisco Vel\u00e1zquez, Kristian Lyngst\u00f8l, Tollef Fog Heen, and J\u00e9r\u00f4me Renard. 2016. The Varnish Book for Varnish 4.0. Varnish Software AS."},{"key":"e_1_3_2_69_2","volume-title":"USENIX ATC","author":"Waldspurger Carl","year":"2017","unstructured":"Carl Waldspurger, Trausti Saemundsson, Irfan Ahmad, and Nohhyun Park. 2017. Cache modeling and optimization using miniature simulations. In USENIX ATC."},{"key":"e_1_3_2_70_2","volume-title":"USENIX HotCloud","author":"Wang Rui","year":"2013","unstructured":"Rui Wang, Christopher Conrad, and Sam Shah. 2013. Using set cover to optimize a large-scale low latency distributed graph. In USENIX HotCloud."},{"key":"e_1_3_2_71_2","volume-title":"USENIX ATC","author":"Wu Xingbo","year":"2015","unstructured":"Xingbo Wu, Yuehai Xu, Zili Shao, and Song Jiang. 2015. LSM-trie: An LSM-tree-based ultra-large key-value store for small data items. In USENIX ATC."},{"issue":"4","key":"e_1_3_2_72_2","first-page":"301","article-title":"Bluecache: A scalable distributed flash-based key-value store","volume":"10","author":"Xu Shuotao","year":"2016","unstructured":"Shuotao Xu, Sungjin Lee, Sang-Woo Jun, Ming Liu, Jamey Hicks, et\u00a0al. 2016. Bluecache: A scalable distributed flash-based key-value store. VLDB 10, 4 (2016), 301\u2013312.","journal-title":"VLDB"},{"key":"e_1_3_2_73_2","doi-asserted-by":"publisher","DOI":"10.1145\/3468521"},{"key":"e_1_3_2_74_2","volume-title":"USENIX OSDI","author":"Yang Juncheng","year":"2020","unstructured":"Juncheng Yang, Yao Yue, and Rashmi Vinayak. 2020. A large scale analysis of hundreds of in-memory cache clusters at Twitter. In USENIX OSDI."},{"key":"e_1_3_2_75_2","volume-title":"USENIX NSDI","author":"Yang Juncheng","year":"2021","unstructured":"Juncheng Yang, Yao Yue, and Rashmi Vinayak. 2021. Segcache: A memory-efficient and scalable in-memory key-value cache for small objects. In USENIX NSDI."},{"key":"e_1_3_2_76_2","article-title":"Taming Tail Latency and Achieving Predictability","author":"Yue Yao","year":"2020","unstructured":"Yao Yue. 2020. Taming Tail Latency and Achieving Predictability. https:\/\/twitter.github.io\/pelikan\/2020\/benchmark-adq.html.","journal-title":"https:\/\/twitter.github.io\/pelikan\/2020\/benchmark-adq.html"}],"container-title":["ACM Transactions on Storage"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3542928","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3542928","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T19:02:22Z","timestamp":1750186942000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3542928"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,8,24]]},"references-count":75,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2022,8,31]]}},"alternative-id":["10.1145\/3542928"],"URL":"https:\/\/doi.org\/10.1145\/3542928","relation":{},"ISSN":["1553-3077","1553-3093"],"issn-type":[{"type":"print","value":"1553-3077"},{"type":"electronic","value":"1553-3093"}],"subject":[],"published":{"date-parts":[[2022,8,24]]},"assertion":[{"value":"2022-02-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-05-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-08-24","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}