{"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/collective-mind-part-ii-towards-performance","title":"Collective Mind, Part II: Towards Performance- and Cost-Aware Software Engineering as a Natural Science","arxiv_id":"1506.06256","date":"2015-06-20","proceeding":null,"authors":["Grigori Fursin","Abdul Memon","Christophe Guillon","Anton Lokhmotov"],"abstract":"Nowadays, engineers have to develop software often without even knowing which\nhardware it will eventually run on in numerous mobile phones, tablets,\ndesktops, laptops, data centers, supercomputers and cloud services.\nUnfortunately, optimizing compilers are not keeping pace with ever increasing\ncomplexity of computer systems anymore and may produce severely underperforming\nexecutable codes while wasting expensive resources and energy.\n  We present our practical and collaborative solution to this problem via\nlight-weight wrappers around any software piece when more than one\nimplementation or optimization choice available. These wrappers are connected\nwith a public Collective Mind autotuning infrastructure and repository of\nknowledge (c-mind.org/repo) to continuously monitor various important\ncharacteristics of these pieces (computational species) across numerous\nexisting hardware configurations together with randomly selected optimizations.\nSimilar to natural sciences, we can now continuously track winning solutions\n(optimizations for a given hardware) that minimize all costs of a computation\n(execution time, energy spent, code size, failures, memory and storage\nfootprint, optimization time, faults, contentions, inaccuracy and so on) of a\ngiven species on a Pareto frontier along with any unexpected behavior. The\ncommunity can then collaboratively classify solutions, prune redundant ones,\nand correlate them with various features of software, its inputs (data sets)\nand used hardware either manually or using powerful predictive analytics\ntechniques. Our approach can then help create a large, realistic, diverse,\nrepresentative, and continuously evolving benchmark with related optimization\nknowledge while gradually covering all possible software and hardware to be\nable to predict best optimizations and improve compilers and hardware depending\non usage scenarios and requirements.","url_abs":"http://arxiv.org/abs/1506.06256v1","url_pdf":"http://arxiv.org/pdf/1506.06256v1.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":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-analytics","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-autotuning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-caffe2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-clsmith","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-crowdtuning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-crowdtuning-platforms","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-dev-compilers","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-docker","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-env","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-math","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-mxnet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"mxnet","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-tensorflow","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-wa","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ck-web","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/crowdsource-experiments-using-android-devices","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/ctuning-programs","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/reproduce-adapt16","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/reproduce-ck-paper","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/reproduce-ck-paper-large-experiments","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/ctuning/reproduce-milepost-project","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/dividiti/ck-caffe","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"collective-mind-part-ii-towards-performance","repo_url":"https://github.com/dividiti/crowdsource-video-experiments-on-android","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","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}