[{"@context":"https:\/\/schema.org\/","@type":"BlogPosting","@id":"https:\/\/blog.terabox.com\/insights\/python-performance-concurrency-memory-model#BlogPosting","mainEntityOfPage":"https:\/\/blog.terabox.com\/insights\/python-performance-concurrency-memory-model","headline":"Python Performance: Why We Need a Formal Memory Model","name":"Python Performance: Why We Need a Formal Memory Model","description":"\ud83d\udcfa Today’s recommended deep-dive video: https:\/\/www.youtube.com\/watch?v=03DswsNUBdQ The Speed Trap: How Faster Python Might Be Breaking Your CodeThe Double-Edged Sword of PerformanceThe Refactoring TrapThe Necessity of a Formal Memory ModelKey TakeawaysQ&A The Speed Trap: How Faster Python Might Be Breaking Your... ","datePublished":"2026-08-18","dateModified":"2026-08-18","author":{"@type":"Person","@id":"https:\/\/blog.terabox.com\/author\/flextech-admin\/#Person","name":"flextech-admin","url":"https:\/\/blog.terabox.com\/author\/flextech-admin\/","image":{"@type":"ImageObject","@id":"https:\/\/secure.gravatar.com\/avatar\/ad516503a11cd5ca435acc9bb6523536?s=150&d=mm&r=gforcedefault=1","url":"https:\/\/secure.gravatar.com\/avatar\/ad516503a11cd5ca435acc9bb6523536?s=150&d=mm&r=gforcedefault=1","height":96,"width":96}},"publisher":{"@type":"Organization","name":"terabox","logo":{"@type":"ImageObject","@id":"http:\/\/blog.terabox.com\/wp-content\/uploads\/2021\/11\/logo\u4ea7\u54c1\u540d-\u7ad6\u7248.png","url":"http:\/\/blog.terabox.com\/wp-content\/uploads\/2021\/11\/logo\u4ea7\u54c1\u540d-\u7ad6\u7248.png","width":900,"height":900}},"image":{"@type":"ImageObject","@id":"https:\/\/img.youtube.com\/vi\/03DswsNUBdQ\/maxresdefault.jpg","url":"https:\/\/img.youtube.com\/vi\/03DswsNUBdQ\/maxresdefault.jpg","height":"","width":""},"url":"https:\/\/blog.terabox.com\/insights\/python-performance-concurrency-memory-model","video":{"@context":"http:\/\/schema.org\/","@type":"VideoObject","@id":"https:\/\/www.youtube.com\/watch?v=03DswsNUBdQ#VideoObject","contentUrl":"https:\/\/www.youtube.com\/watch?v=03DswsNUBdQ","name":"[Sponsor Invited Talks'25] Python, Is It Being Killed by Incremental Improvements?","description":"Python, Is It Being Killed by Incremental Improvements? (Video, Sponsorship Invited Talks 2025)\nStefan Marr\n(Johannes Kepler University Linz)\n\nAbstract: Over the past years, two major players invested into the future of Python. Microsoft\u2019s Faster CPython team is pushed ahead with impressive performance improvements for the CPython interpreter, which has gotten at least 2x faster since Python 3.9. They also have a baseline JIT compiler for CPython, too. At the same time, Meta is worked hard on making free-threaded Python a reality to bring classic shared-memory multithreading to Python, without being limited by the still standard Global Interpreter Lock, which prevents true parallelism.\nBoth projects deliver major improvements to Python, and the wider ecosystem. So, it\u2019s all great, or is it?\nIn this talk, I\u2019ll discuss some of the aspects the Python core developers and wider community seem to not regard with the same urgency as I would hope for. Concurrency makes me scared, and I strongly believe the Python ecosystem should be scared, too, or look forward to the 2030s being \u201cPython\u2019s Decade of Concurrency Bugs\u201d. We\u2019ll start out reviewing some of the changes in observable language semantics between Python 3.9 and today, discuss their implications, and because of course I have some old ideas lying around, try to propose a way fordward. In practice though, this isn\u2019t a small well-defined research project. So, I hope I can inspire some of you to follow me down the rabbit hole of Python\u2019s free-threaded future.\n\n\nPresentation at the Sponsorship Invited Talks 2025 forum, Oct 18, 2025, https:\/\/conf.researchr.org\/home\/icfp-splash-2025\/industry-forum-2025\nSponsored by ACM SIGPLAN.","thumbnailUrl":["https:\/\/i.ytimg.com\/vi\/03DswsNUBdQ\/default.jpg","https:\/\/i.ytimg.com\/vi\/03DswsNUBdQ\/mqdefault.jpg","https:\/\/i.ytimg.com\/vi\/03DswsNUBdQ\/hqdefault.jpg","https:\/\/i.ytimg.com\/vi\/03DswsNUBdQ\/sddefault.jpg","https:\/\/i.ytimg.com\/vi\/03DswsNUBdQ\/maxresdefault.jpg"],"uploadDate":"2025-12-05T13:32:46+00:00","duration":"PT29M3S","embedUrl":"https:\/\/www.youtube.com\/embed\/03DswsNUBdQ","publisher":{"@type":"Organization","@id":"https:\/\/www.youtube.com\/channel\/UCwG9512Wm7jSS6Iqshz4Dpg#Organization","url":"https:\/\/www.youtube.com\/channel\/UCwG9512Wm7jSS6Iqshz4Dpg","name":"ACM SIGPLAN","description":"Special Interest Group on Programming Languages\nThe ACM Special Interest Group on Programming Languages (SIGPLAN) explores programming language concepts and tools, focusing on design, implementation, practice, and theory. Its members are programming language developers, educators, implementers, researchers, theoreticians, and users.\n\nAll videos published on this channel are Copyright ACM. \n\nPlease email SIGPLAN-AV for video take-down requests or corrections.","logo":{"url":"https:\/\/yt3.ggpht.com\/ytc\/AIdro_mB1glk2WSnakJ9VmIADfpyj122SV1iL5zw1rMBEQFIqQ=s800-c-k-c0x00ffffff-no-rj","width":800,"height":800,"@type":"ImageObject","@id":"https:\/\/www.youtube.com\/watch?v=03DswsNUBdQ#VideoObject_publisher_logo_ImageObject"}},"potentialAction":{"@type":"SeekToAction","@id":"https:\/\/www.youtube.com\/watch?v=03DswsNUBdQ#VideoObject_potentialAction","target":"https:\/\/www.youtube.com\/watch?v=03DswsNUBdQ&t={seek_to_second_number}","startOffset-input":"required name=seek_to_second_number"},"interactionStatistic":[[{"@type":"InteractionCounter","@id":"https:\/\/www.youtube.com\/watch?v=03DswsNUBdQ#VideoObject_interactionStatistic_WatchAction","interactionType":{"@type":"WatchAction"},"userInteractionCount":1235}],{"@type":"InteractionCounter","@id":"https:\/\/www.youtube.com\/watch?v=03DswsNUBdQ#VideoObject_interactionStatistic_LikeAction","interactionType":{"@type":"LikeAction"},"userInteractionCount":36}]},"about":["Insights","\u300eEnglish\u300f"],"wordCount":1516,"keywords":["Machine Learning","sync"]},{"@context":"https:\/\/schema.org\/","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Insights","item":"https:\/\/blog.terabox.com\/insights\/#breadcrumbitem"},{"@type":"ListItem","position":2,"name":"Python Performance: Why We Need a Formal Memory Model","item":"https:\/\/blog.terabox.com\/insights\/python-performance-concurrency-memory-model#breadcrumbitem"}]}]