{"main_file_link":[{"open_access":"1","url":"https://www.usenix.org/system/files/conference/atc18/atc18-arbel-raviv.pdf"}],"abstract":[{"lang":"eng","text":"Many systems rely on optimistic concurrent search trees for multi-core scalability. In principle, optimistic trees have a simple performance story: searches are read-only and so run in parallel, with writes to shared memory occurring only when modifying the data structure. However, this paper shows that in practice, obtaining the full performance benefits of optimistic search trees is not so simple.\r\n\r\nWe focus on optimistic binary search trees (BSTs) and perform a detailed performance analysis of 10 state-of-the-art BSTs on large scale x86-64 hardware, using both microbenchmarks and an in-memory database system. We find and explain significant unexpected performance differences between BSTs with similar tree structure and search implementations, which we trace to subtle performance-degrading interactions of BSTs with systems software and hardware subsystems. We further derive a prescriptive approach to avoid this performance degradation, as well as algorithmic insights on optimistic BST design. Our work underlines the gap between the theory and practice of multi-core performance, and calls for further research to help bridge this gap."}],"_id":"7272","language":[{"iso":"eng"}],"department":[{"_id":"DaAl"}],"citation":{"ieee":"M. Arbel-Raviv, T. A. Brown, and A. Morrison, “Getting to the root of concurrent binary search tree performance,” in <i>Proceedings of the 2018 USENIX Annual Technical Conference</i>, Boston, MA, United States, 2020, pp. 295–306.","mla":"Arbel-Raviv, Maya, et al. “Getting to the Root of Concurrent Binary Search Tree Performance.” <i>Proceedings of the 2018 USENIX Annual Technical Conference</i>, USENIX Association, 2020, pp. 295–306.","short":"M. Arbel-Raviv, T.A. Brown, A. Morrison, in:, Proceedings of the 2018 USENIX Annual Technical Conference, USENIX Association, 2020, pp. 295–306.","chicago":"Arbel-Raviv, Maya, Trevor A Brown, and Adam Morrison. “Getting to the Root of Concurrent Binary Search Tree Performance.” In <i>Proceedings of the 2018 USENIX Annual Technical Conference</i>, 295–306. USENIX Association, 2020.","ista":"Arbel-Raviv M, Brown TA, Morrison A. 2020. Getting to the root of concurrent binary search tree performance. Proceedings of the 2018 USENIX Annual Technical Conference. USENIX: Annual Technical Conference, 295–306.","apa":"Arbel-Raviv, M., Brown, T. A., &#38; Morrison, A. (2020). Getting to the root of concurrent binary search tree performance. In <i>Proceedings of the 2018 USENIX Annual Technical Conference</i> (pp. 295–306). Boston, MA, United States: USENIX Association.","ama":"Arbel-Raviv M, Brown TA, Morrison A. Getting to the root of concurrent binary search tree performance. In: <i>Proceedings of the 2018 USENIX Annual Technical Conference</i>. USENIX Association; 2020:295-306."},"conference":{"start_date":"2018-07-11","name":"USENIX: Annual Technical Conference","end_date":"2018-07-13","location":"Boston, MA, United States"},"date_published":"2020-01-01T00:00:00Z","publication_status":"published","date_updated":"2021-01-11T15:25:48Z","publication_identifier":{"isbn":["9781939133021"]},"article_processing_charge":"No","month":"01","author":[{"first_name":"Maya","last_name":"Arbel-Raviv","full_name":"Arbel-Raviv, Maya"},{"last_name":"Brown","first_name":"Trevor A","id":"3569F0A0-F248-11E8-B48F-1D18A9856A87","full_name":"Brown, Trevor A"},{"last_name":"Morrison","first_name":"Adam","full_name":"Morrison, Adam"}],"day":"01","publication":"Proceedings of the 2018 USENIX Annual Technical Conference","page":"295-306","type":"conference","project":[{"name":"NSERC Postdoctoral fellowship","_id":"26450934-B435-11E9-9278-68D0E5697425"}],"ddc":["000"],"status":"public","quality_controlled":"1","title":"Getting to the root of concurrent binary search tree performance","scopus_import":"1","year":"2020","publisher":"USENIX Association","user_id":"2DF688A6-F248-11E8-B48F-1D18A9856A87","date_created":"2020-01-14T07:27:08Z","oa_version":"Published Version","oa":1}