{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/dimmwitted-a-study-of-main-memory-statistical","title":"DimmWitted: A Study of Main-Memory Statistical Analytics","arxiv_id":"1403.7550","date":"2014-03-28","proceeding":null,"authors":["Ce Zhang","Christopher Ré"],"abstract":"We perform the first study of the tradeoff space of access methods and\nreplication to support statistical analytics using first-order methods executed\nin the main memory of a Non-Uniform Memory Access (NUMA) machine. Statistical\nanalytics systems differ from conventional SQL-analytics in the amount and\ntypes of memory incoherence they can tolerate. Our goal is to understand\ntradeoffs in accessing the data in row- or column-order and at what granularity\none should share the model and data for a statistical task. We study this new\ntradeoff space, and discover there are tradeoffs between hardware and\nstatistical efficiency. We argue that our tradeoff study may provide valuable\ninformation for designers of analytics engines: for each system we consider,\nour prototype engine can run at least one popular task at least 100x faster. We\nconduct our study across five architectures using popular models including\nSVMs, logistic regression, Gibbs sampling, and neural networks.","url_abs":"http://arxiv.org/abs/1403.7550v3","url_pdf":"http://arxiv.org/pdf/1403.7550v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"dimmwitted-a-study-of-main-memory-statistical","repo_url":"https://github.com/HazyResearch/CaffeConTroll","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}